{"id":1507,"date":"2026-09-18T01:10:54","date_gmt":"2026-09-17T16:10:54","guid":{"rendered":"https:\/\/localmodelwatch.tsuchitsuchi.com\/2026\/09\/18\/tencent-releases-wevisdoc-document-parsing-models\/"},"modified":"2026-11-03T16:08:35","modified_gmt":"2026-11-03T07:08:35","slug":"tencent-releases-wevisdoc-document-parsing-models","status":"publish","type":"post","link":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/18\/tencent-releases-wevisdoc-document-parsing-models\/","title":{"rendered":"Tencent Releases WeVisDoc-2B and WeVisDoc-4B Document Parsing Models"},"content":{"rendered":"<p><!-- lmw:facts --><\/p>\n<h2>At a Glance<\/h2>\n<div class=\"lmw-table-scroll\" tabindex=\"0\" style=\"overflow-x:auto;-webkit-overflow-scrolling:touch;max-width:100%;\">\n<table style=\"width:max-content;min-width:100%;border-collapse:collapse;\">\n<thead>\n<tr>\n<th>Item<\/th>\n<th>Value<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Repository<\/td>\n<td><a href=\"https:\/\/huggingface.co\/tencent\/WeVisDoc-4B\">tencent\/WeVisDoc-4B<\/a><\/td>\n<\/tr>\n<tr>\n<td>Family guide<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/model-tencent-wevisdoc-en\/\">WeVisDoc guide (1 articles)<\/a><\/td>\n<\/tr>\n<tr>\n<td>Publisher guide<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/publisher-tencent-en\/\">Tencent: models and licenses<\/a><\/td>\n<\/tr>\n<tr>\n<td>Published<\/td>\n<td>2026-09-16<\/td>\n<\/tr>\n<tr>\n<td>License<\/td>\n<td>apache-2.0<\/td>\n<\/tr>\n<tr>\n<td>Formats<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/format-safetensors-en\/\">safetensors<\/a><\/td>\n<\/tr>\n<tr>\n<td>Paper<\/td>\n<td><a href=\"https:\/\/arxiv.org\/abs\/2609.20423\">arXiv:2609.20423<\/a><\/td>\n<\/tr>\n<tr>\n<td>Source type<\/td>\n<td>Primary source (the publisher itself)<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p><em>Values determined by this site&#8217;s code when the information was collected. Dates are JST.<\/em><\/p>\n<p><!-- \/lmw:facts --><\/p>\n<p><!-- lmw:lab-summary --><\/p>\n<p><strong>What we checked ourselves<\/strong><\/p>\n<ul>\n<li>Ran <code>WeVisDoc-4B.Q4_K_M.gguf<\/code> (Q4_K_M) on our CPU-only server (Neoverse-N1, 3 threads, llama.cpp b11223): generation at 8.3 tokens\/s (a CPU figure, not a GPU speed).<\/li>\n<li>Its answers to our fixed Japanese questions are shown verbatim below.<\/li>\n<li>The same Japanese text takes about as many tokens as with the Qwen3 tokenizer.<\/li>\n<\/ul>\n<p>Details and conditions are in \u201cOur Own Measurements\u201d below.<\/p>\n<p><!-- \/lmw:lab-summary --><\/p>\n<h2>Overview<\/h2>\n<p>Tencent has released &#8220;WeVisDoc-2B&#8221; and &#8220;WeVisDoc-4B&#8221;, end-to-end document parsing models that extract structured text from page images.<\/p>\n<p>These models are fine-tuned versions based on &#8220;Qwen3-VL-2B-Instruct&#8221; and &#8220;Qwen3-VL-4B-Instruct&#8221; respectively, taking page images as input and outputting structured Markdown containing LaTeX-format mathematical formulas and HTML-format tables.<\/p>\n<h2>Specifications<\/h2>\n<ul>\n<li>Architecture: Qwen3VLForConditionalGeneration (qwen3_vl)<\/li>\n<li>Base Models: Qwen3-VL-2B-Instruct, Qwen3-VL-4B-Instruct<\/li>\n<li>License: apache-2.0<\/li>\n<\/ul>\n<h2>Performance<\/h2>\n<p>The measurement results of document parsing benchmarks provided by the publisher are as follows. Evaluation values represent the average of three inference runs.<\/p>\n<h3>OmniDocBench v1.6<\/h3>\n<div class=\"lmw-table-scroll\" tabindex=\"0\" style=\"overflow-x:auto;-webkit-overflow-scrolling:touch;max-width:100%;\">\n<p class=\"lmw-table-hint\" style=\"margin:0 0 4px;font-size:0.85em;opacity:0.7;\">\u2192 Scroll horizontally to see all columns<\/p>\n<table style=\"width:max-content;min-width:100%;border-collapse:collapse;\">\n<thead>\n<tr>\n<th>Model<\/th>\n<th style=\"text-align: right;\">Params<\/th>\n<th style=\"text-align: right;\">Overall \u2191<\/th>\n<th style=\"text-align: right;\">TextEdit \u2193<\/th>\n<th style=\"text-align: right;\">FormulaCDM \u2191<\/th>\n<th style=\"text-align: right;\">TableTEDS \u2191<\/th>\n<th style=\"text-align: right;\">TableTEDS_S \u2191<\/th>\n<th style=\"text-align: right;\">ROEdit \u2193<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Nanonets-OCR2*<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">83.20<\/td>\n<td style=\"text-align: right;\">0.108<\/td>\n<td style=\"text-align: right;\">80.35<\/td>\n<td style=\"text-align: right;\">80.10<\/td>\n<td style=\"text-align: right;\">85.26<\/td>\n<td style=\"text-align: right;\">0.211<\/td>\n<\/tr>\n<tr>\n<td>OCRFlux-3B*<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">83.31<\/td>\n<td style=\"text-align: right;\">0.126<\/td>\n<td style=\"text-align: right;\">88.75<\/td>\n<td style=\"text-align: right;\">73.78<\/td>\n<td style=\"text-align: right;\">77.98<\/td>\n<td style=\"text-align: right;\">0.217<\/td>\n<\/tr>\n<tr>\n<td>POINTS-Reader<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">83.37<\/td>\n<td style=\"text-align: right;\">0.096<\/td>\n<td style=\"text-align: right;\">85.72<\/td>\n<td style=\"text-align: right;\">73.98<\/td>\n<td style=\"text-align: right;\">77.40<\/td>\n<td style=\"text-align: right;\">0.198<\/td>\n<\/tr>\n<tr>\n<td>Nanonets-OCR-s<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">83.61<\/td>\n<td style=\"text-align: right;\">0.108<\/td>\n<td style=\"text-align: right;\">81.46<\/td>\n<td style=\"text-align: right;\">80.18<\/td>\n<td style=\"text-align: right;\">84.51<\/td>\n<td style=\"text-align: right;\">0.213<\/td>\n<\/tr>\n<tr>\n<td>olmOCR-2-7B*<\/td>\n<td style=\"text-align: right;\">7B<\/td>\n<td style=\"text-align: right;\">85.51<\/td>\n<td style=\"text-align: right;\">0.106<\/td>\n<td style=\"text-align: right;\">88.84<\/td>\n<td style=\"text-align: right;\">78.32<\/td>\n<td style=\"text-align: right;\">82.81<\/td>\n<td style=\"text-align: right;\">0.223<\/td>\n<\/tr>\n<tr>\n<td>olmOCR<\/td>\n<td style=\"text-align: right;\">7B<\/td>\n<td style=\"text-align: right;\">85.74<\/td>\n<td style=\"text-align: right;\">0.139<\/td>\n<td style=\"text-align: right;\">88.10<\/td>\n<td style=\"text-align: right;\">83.00<\/td>\n<td style=\"text-align: right;\">87.17<\/td>\n<td style=\"text-align: right;\">0.216<\/td>\n<\/tr>\n<tr>\n<td>DeepSeek-OCR*<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">86.31<\/td>\n<td style=\"text-align: right;\">0.077<\/td>\n<td style=\"text-align: right;\">84.71<\/td>\n<td style=\"text-align: right;\">81.87<\/td>\n<td style=\"text-align: right;\">86.07<\/td>\n<td style=\"text-align: right;\">0.171<\/td>\n<\/tr>\n<tr>\n<td>OCRVerse<\/td>\n<td style=\"text-align: right;\">4B<\/td>\n<td style=\"text-align: right;\">88.60<\/td>\n<td style=\"text-align: right;\">0.063<\/td>\n<td style=\"text-align: right;\">89.61<\/td>\n<td style=\"text-align: right;\">82.44<\/td>\n<td style=\"text-align: right;\">86.27<\/td>\n<td style=\"text-align: right;\">0.163<\/td>\n<\/tr>\n<tr>\n<td>UniRec-0.1B*<\/td>\n<td style=\"text-align: right;\">0.1B<\/td>\n<td style=\"text-align: right;\">88.91<\/td>\n<td style=\"text-align: right;\">0.088<\/td>\n<td style=\"text-align: right;\">92.14<\/td>\n<td style=\"text-align: right;\">83.40<\/td>\n<td style=\"text-align: right;\">86.79<\/td>\n<td style=\"text-align: right;\">0.146<\/td>\n<\/tr>\n<tr>\n<td>DeepSeek-OCR 2<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">90.25<\/td>\n<td style=\"text-align: right;\">0.050<\/td>\n<td style=\"text-align: right;\">91.84<\/td>\n<td style=\"text-align: right;\">83.89<\/td>\n<td style=\"text-align: right;\">87.75<\/td>\n<td style=\"text-align: right;\">0.144<\/td>\n<\/tr>\n<tr>\n<td>dots.ocr<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">90.77<\/td>\n<td style=\"text-align: right;\">0.048<\/td>\n<td style=\"text-align: right;\">89.95<\/td>\n<td style=\"text-align: right;\">87.18<\/td>\n<td style=\"text-align: right;\">90.58<\/td>\n<td style=\"text-align: right;\">0.138<\/td>\n<\/tr>\n<tr>\n<td>FD-RL*<\/td>\n<td style=\"text-align: right;\">4B<\/td>\n<td style=\"text-align: right;\">91.21<\/td>\n<td style=\"text-align: right;\">0.055<\/td>\n<td style=\"text-align: right;\">92.92<\/td>\n<td style=\"text-align: right;\">86.22<\/td>\n<td style=\"text-align: right;\">90.92<\/td>\n<td style=\"text-align: right;\">0.145<\/td>\n<\/tr>\n<tr>\n<td>HunyuanOCR<\/td>\n<td style=\"text-align: right;\">1B<\/td>\n<td style=\"text-align: right;\">92.03<\/td>\n<td style=\"text-align: right;\">0.048<\/td>\n<td style=\"text-align: right;\">88.60<\/td>\n<td style=\"text-align: right;\">92.37<\/td>\n<td style=\"text-align: right;\">93.99<\/td>\n<td style=\"text-align: right;\">0.138<\/td>\n<\/tr>\n<tr>\n<td>dots.mocr*<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">92.57<\/td>\n<td style=\"text-align: right;\">0.042<\/td>\n<td style=\"text-align: right;\">92.09<\/td>\n<td style=\"text-align: right;\">89.78<\/td>\n<td style=\"text-align: right;\">92.92<\/td>\n<td style=\"text-align: right;\">0.133<\/td>\n<\/tr>\n<tr>\n<td>FireRed-OCR<\/td>\n<td style=\"text-align: right;\">2B<\/td>\n<td style=\"text-align: right;\">93.26<\/td>\n<td style=\"text-align: right;\">0.037<\/td>\n<td style=\"text-align: right;\">95.44<\/td>\n<td style=\"text-align: right;\">88.04<\/td>\n<td style=\"text-align: right;\">91.06<\/td>\n<td style=\"text-align: right;\">0.131<\/td>\n<\/tr>\n<tr>\n<td>Logics-Parsing-v2<\/td>\n<td style=\"text-align: right;\">4B<\/td>\n<td style=\"text-align: right;\">93.33<\/td>\n<td style=\"text-align: right;\">0.041<\/td>\n<td style=\"text-align: right;\">95.65<\/td>\n<td style=\"text-align: right;\">88.42<\/td>\n<td style=\"text-align: right;\">91.98<\/td>\n<td style=\"text-align: right;\">0.137<\/td>\n<\/tr>\n<tr>\n<td>Qianfan-OCR<\/td>\n<td style=\"text-align: right;\">4B<\/td>\n<td style=\"text-align: right;\">93.90<\/td>\n<td style=\"text-align: right;\">0.040<\/td>\n<td style=\"text-align: right;\">95.08<\/td>\n<td style=\"text-align: right;\">90.53<\/td>\n<td style=\"text-align: right;\">93.31<\/td>\n<td style=\"text-align: right;\">0.130<\/td>\n<\/tr>\n<tr>\n<td>Unlimited-OCR<\/td>\n<td style=\"text-align: right;\">3B-A0.5B<\/td>\n<td style=\"text-align: right;\">93.92<\/td>\n<td style=\"text-align: right;\">0.042<\/td>\n<td style=\"text-align: right;\">95.79<\/td>\n<td style=\"text-align: right;\">90.16<\/td>\n<td style=\"text-align: right;\">93.32<\/td>\n<td style=\"text-align: right;\">0.129<\/td>\n<\/tr>\n<tr>\n<td>HunyuanOCR-1.5<\/td>\n<td style=\"text-align: right;\">1B<\/td>\n<td style=\"text-align: right;\">94.74<\/td>\n<td style=\"text-align: right;\">0.039<\/td>\n<td style=\"text-align: right;\">94.50<\/td>\n<td style=\"text-align: right;\">93.67<\/td>\n<td style=\"text-align: right;\">94.71<\/td>\n<td style=\"text-align: right;\">0.129<\/td>\n<\/tr>\n<tr>\n<td>WeVisDoc-2B<\/td>\n<td style=\"text-align: right;\">2B<\/td>\n<td style=\"text-align: right;\">95.06<\/td>\n<td style=\"text-align: right;\">0.038<\/td>\n<td style=\"text-align: right;\">95.94<\/td>\n<td style=\"text-align: right;\">93.03<\/td>\n<td style=\"text-align: right;\">95.26<\/td>\n<td style=\"text-align: right;\">0.130<\/td>\n<\/tr>\n<tr>\n<td>WeVisDoc-4B<\/td>\n<td style=\"text-align: right;\">4B<\/td>\n<td style=\"text-align: right;\">95.38<\/td>\n<td style=\"text-align: right;\">0.036<\/td>\n<td style=\"text-align: right;\">96.81<\/td>\n<td style=\"text-align: right;\">92.95<\/td>\n<td style=\"text-align: right;\">95.34<\/td>\n<td style=\"text-align: right;\">0.125<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<h3>PureDocBench<\/h3>\n<div class=\"lmw-table-scroll\" tabindex=\"0\" style=\"overflow-x:auto;-webkit-overflow-scrolling:touch;max-width:100%;\">\n<p class=\"lmw-table-hint\" style=\"margin:0 0 4px;font-size:0.85em;opacity:0.7;\">\u2192 Scroll horizontally to see all columns<\/p>\n<table style=\"width:max-content;min-width:100%;border-collapse:collapse;\">\n<thead>\n<tr>\n<th>Model<\/th>\n<th style=\"text-align: right;\">Params<\/th>\n<th style=\"text-align: right;\">Avg\u2083 \u2191<\/th>\n<th style=\"text-align: right;\">Clean Overall \u2191<\/th>\n<th style=\"text-align: right;\">Digital Degraded Overall \u2191<\/th>\n<th style=\"text-align: right;\">Real Degraded Overall \u2191<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>OCRFlux-3B<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">42.06<\/td>\n<td style=\"text-align: right;\">47.14<\/td>\n<td style=\"text-align: right;\">41.82<\/td>\n<td style=\"text-align: right;\">37.21<\/td>\n<\/tr>\n<tr>\n<td>DeepSeek-OCR<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">46.98<\/td>\n<td style=\"text-align: right;\">53.50<\/td>\n<td style=\"text-align: right;\">46.95<\/td>\n<td style=\"text-align: right;\">40.48<\/td>\n<\/tr>\n<tr>\n<td>UniRec-0.1B<\/td>\n<td style=\"text-align: right;\">0.1B<\/td>\n<td style=\"text-align: right;\">48.59<\/td>\n<td style=\"text-align: right;\">58.91<\/td>\n<td style=\"text-align: right;\">52.42<\/td>\n<td style=\"text-align: right;\">34.44<\/td>\n<\/tr>\n<tr>\n<td>POINTS-Reader*<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">49.24<\/td>\n<td style=\"text-align: right;\">53.78<\/td>\n<td style=\"text-align: right;\">51.24<\/td>\n<td style=\"text-align: right;\">42.69<\/td>\n<\/tr>\n<tr>\n<td>DeepSeek-OCR-2<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">49.51<\/td>\n<td style=\"text-align: right;\">55.53<\/td>\n<td style=\"text-align: right;\">49.41<\/td>\n<td style=\"text-align: right;\">43.60<\/td>\n<\/tr>\n<tr>\n<td>Qianfan-OCR<\/td>\n<td style=\"text-align: right;\">4B<\/td>\n<td style=\"text-align: right;\">51.04<\/td>\n<td style=\"text-align: right;\">57.22<\/td>\n<td style=\"text-align: right;\">50.85<\/td>\n<td style=\"text-align: right;\">45.06<\/td>\n<\/tr>\n<tr>\n<td>olmOCR-7B<\/td>\n<td style=\"text-align: right;\">7B<\/td>\n<td style=\"text-align: right;\">55.90<\/td>\n<td style=\"text-align: right;\">62.56<\/td>\n<td style=\"text-align: right;\">57.84<\/td>\n<td style=\"text-align: right;\">47.30<\/td>\n<\/tr>\n<tr>\n<td>Nanonets-OCR2<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">58.36<\/td>\n<td style=\"text-align: right;\">64.83<\/td>\n<td style=\"text-align: right;\">61.23<\/td>\n<td style=\"text-align: right;\">49.03<\/td>\n<\/tr>\n<tr>\n<td>HunyuanOCR<\/td>\n<td style=\"text-align: right;\">1B<\/td>\n<td style=\"text-align: right;\">60.56<\/td>\n<td style=\"text-align: right;\">65.61<\/td>\n<td style=\"text-align: right;\">61.49<\/td>\n<td style=\"text-align: right;\">54.58<\/td>\n<\/tr>\n<tr>\n<td>Unlimited-OCR*<\/td>\n<td style=\"text-align: right;\">3B-A0.5B<\/td>\n<td style=\"text-align: right;\">62.76<\/td>\n<td style=\"text-align: right;\">71.28<\/td>\n<td style=\"text-align: right;\">63.62<\/td>\n<td style=\"text-align: right;\">53.39<\/td>\n<\/tr>\n<tr>\n<td>olmOCR-2-7B<\/td>\n<td style=\"text-align: right;\">7B<\/td>\n<td style=\"text-align: right;\">63.78<\/td>\n<td style=\"text-align: right;\">69.36<\/td>\n<td style=\"text-align: right;\">65.87<\/td>\n<td style=\"text-align: right;\">56.10<\/td>\n<\/tr>\n<tr>\n<td>dots.ocr<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">64.55<\/td>\n<td style=\"text-align: right;\">72.01<\/td>\n<td style=\"text-align: right;\">65.95<\/td>\n<td style=\"text-align: right;\">55.68<\/td>\n<\/tr>\n<tr>\n<td>Nanonets-OCR-s*<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">65.37<\/td>\n<td style=\"text-align: right;\">71.26<\/td>\n<td style=\"text-align: right;\">66.56<\/td>\n<td style=\"text-align: right;\">58.28<\/td>\n<\/tr>\n<tr>\n<td>FireRed-OCR<\/td>\n<td style=\"text-align: right;\">2B<\/td>\n<td style=\"text-align: right;\">65.57<\/td>\n<td style=\"text-align: right;\">70.81<\/td>\n<td style=\"text-align: right;\">68.49<\/td>\n<td style=\"text-align: right;\">57.42<\/td>\n<\/tr>\n<tr>\n<td>HunyuanOCR-1.5*<\/td>\n<td style=\"text-align: right;\">1B<\/td>\n<td style=\"text-align: right;\">68.79<\/td>\n<td style=\"text-align: right;\">73.98<\/td>\n<td style=\"text-align: right;\">70.81<\/td>\n<td style=\"text-align: right;\">61.59<\/td>\n<\/tr>\n<tr>\n<td>OCRVerse<\/td>\n<td style=\"text-align: right;\">4B<\/td>\n<td style=\"text-align: right;\">69.40<\/td>\n<td style=\"text-align: right;\">73.18<\/td>\n<td style=\"text-align: right;\">71.36<\/td>\n<td style=\"text-align: right;\">63.66<\/td>\n<\/tr>\n<tr>\n<td>dots.mocr<\/td>\n<td style=\"text-align: right;\">3B<\/td>\n<td style=\"text-align: right;\">70.39<\/td>\n<td style=\"text-align: right;\">76.27<\/td>\n<td style=\"text-align: right;\">73.16<\/td>\n<td style=\"text-align: right;\">61.73<\/td>\n<\/tr>\n<tr>\n<td>Logics-Parsing-v2<\/td>\n<td style=\"text-align: right;\">4B<\/td>\n<td style=\"text-align: right;\">72.61<\/td>\n<td style=\"text-align: right;\">76.35<\/td>\n<td style=\"text-align: right;\">73.85<\/td>\n<td style=\"text-align: right;\">67.64<\/td>\n<\/tr>\n<tr>\n<td>FD-RL<\/td>\n<td style=\"text-align: right;\">4B<\/td>\n<td style=\"text-align: right;\">73.92<\/td>\n<td style=\"text-align: right;\">78.38<\/td>\n<td style=\"text-align: right;\">76.33<\/td>\n<td style=\"text-align: right;\">67.04<\/td>\n<\/tr>\n<tr>\n<td>WeVisDoc-2B<\/td>\n<td style=\"text-align: right;\">2B<\/td>\n<td style=\"text-align: right;\">73.86<\/td>\n<td style=\"text-align: right;\">79.36<\/td>\n<td style=\"text-align: right;\">76.62<\/td>\n<td style=\"text-align: right;\">65.60<\/td>\n<\/tr>\n<tr>\n<td>WeVisDoc-4B<\/td>\n<td style=\"text-align: right;\">4B<\/td>\n<td style=\"text-align: right;\">75.54<\/td>\n<td style=\"text-align: right;\">79.81<\/td>\n<td style=\"text-align: right;\">77.74<\/td>\n<td style=\"text-align: right;\">69.08<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p>According to the measurement results stated in the model card, WeVisDoc-4B achieves the highest scores among compared end-to-end parsers in both the OmniDocBench v1.6 Overall (95.38) and the PureDocBench 3-track average Avg\u2083 (75.54). It exhibits particularly high figures in OmniDocBench v1.6 FormulaCDM (96.81) and PureDocBench Clean Overall (79.81), demonstrating superior parsing accuracy for mathematical formulas and standard document images. The smaller WeVisDoc-2B is also remarkably capable, recording 95.06 in OmniDocBench v1.6 Overall and outperforming larger and other models such as HunyuanOCR-1.5 and Logics-Parsing-v2. On the other hand, in PureDocBench&#8217;s &#8220;Real Degraded Overall&#8221;, WeVisDoc-2B stays at 65.60, trailing behind FD-RL (67.04) and Logics-Parsing-v2 (67.64), showing a difference in robustness toward real photographed degraded images compared to the larger 4B version.<\/p>\n<h2>Strengths and Use Cases<\/h2>\n<p>WeVisDoc specializes in generating structured Markdown from page images (PNG, JPEG, WebP) of books, research papers, and various documents. It excels not only in converting body text but also in extracting mathematical formulas using LaTeX format and analyzing table structures using HTML format. English (en) and Chinese (zh) are set as tag information, making it suitable for OCR and paper parsing use cases involving multilingual documents.<\/p>\n<p><!-- lmw:hardware --><\/p>\n<h2>Hardware Requirements<\/h2>\n<p><strong>Estimated requirements (calculated by Local Model Watch)<\/strong> \u2014 4.4B parameters<\/p>\n<div class=\"lmw-table-scroll\" tabindex=\"0\" style=\"overflow-x:auto;-webkit-overflow-scrolling:touch;max-width:100%;\">\n<table style=\"width:max-content;min-width:100%;border-collapse:collapse;\">\n<thead>\n<tr>\n<th>Your VRAM<\/th>\n<th>Quantization<\/th>\n<th>File size<\/th>\n<th>Est. memory needed<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>4GB (laptop iGPU \/ phone class)<\/td>\n<td>Q5_K_S<\/td>\n<td>2.9GB<\/td>\n<td>3.5GB<\/td>\n<\/tr>\n<tr>\n<td>8GB (RTX 4060 \/ 3060 Ti, etc.)<\/td>\n<td>Q8_0<\/td>\n<td>4.4GB<\/td>\n<td>5.2GB<\/td>\n<\/tr>\n<tr>\n<td>12GB (RTX 4070 \/ 3060 12GB, etc.)<\/td>\n<td>F16<\/td>\n<td>8.2GB<\/td>\n<td>9.9GB<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p><strong>Inference engine support<\/strong> (architecture name matched against each project&#8217;s own model registry in its source code, checked 2026-11-03): llama.cpp: registered, vLLM: registered, MLX (mlx-lm): registered.<\/p>\n<p><em>Memory estimates add a 20% runtime overhead (KV cache, etc.) to the actual size of the distributed files. Actual usage varies with context length, batch size and inference engine. These figures are computed by this site from file sizes, not published by the model&#8217;s authors. File sizes are measured from the converted build <a href=\"https:\/\/huggingface.co\/mradermacher\/WeVisDoc-4B-GGUF\">mradermacher\/WeVisDoc-4B-GGUF<\/a>. Compare with other models in our <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/vram-guide-en\/\">VRAM quick reference<\/a>. What the quantization names mean: <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/glossary-quantization-en\/\">glossary<\/a>.<\/em><\/p>\n<p><!-- \/lmw:hardware --><\/p>\n<p><!-- lmw:runnability --><\/p>\n<h2>Can You Run It Locally?<\/h2>\n<p>The publisher distributes this model as safetensors.<\/p>\n<p><strong>License \u2014 <code>apache-2.0<\/code> (Commercial use allowed):<\/strong> Permits commercial use, modification and redistribution. Redistribution requires including the license and stating changes; includes a patent grant.<\/p>\n<p><strong>Compression:<\/strong> the Q5_K_S build measures 5.61 bits per weight \u2014 about 35% the size of the original 16-bit weights, calculated by this site from the actual file sizes.<\/p>\n<p><em>Compiled by this site&#8217;s code from the published formats and the license field. License summaries are not legal advice \u2014 check the publisher&#8217;s original terms before relying on them.<\/em><\/p>\n<p><!-- \/lmw:runnability --><\/p>\n<p><!-- lmw:lab --><\/p>\n<h2>Our Own Measurements<\/h2>\n<p>Values we measured ourselves on our server (no GPU) by actually reading and running this model&#8217;s files \u2014 not figures copied from the model card. How we measure, and the results for every model: <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/observations-en\/\">Our Measurements<\/a>.<\/p>\n<h3>What We Found<\/h3>\n<p>When running WeVisDoc-4B solely on a CPU, it successfully handled tasks such as summarizing text, writing emails with polite phrasing, and solving shopping word problems, but stumbled on a knowledge question asking for prefectures without an ocean coastline, resulting in the same line repeating 16 times and hitting the 1,024-token limit. In an environment with Neoverse-N1 and 3 threads using Q4_K_M quantization, the generation speed was 8.31 tokens per second, which translates to 12.1 Japanese characters per second. The peak memory consumption was 6.7GB, and the file size is 2.53GB. In the image reading tests, it correctly answered the busiest day of the week and the correct figure of 260 people from a visitor volume graph, but failed to provide the correct total for weekdays or the correct total amount for a shopping receipt.<\/p>\n<p>The Japanese token efficiency is 688 tokens per 1,000 characters, with a token ratio of 1.26 compared to the same content in English. Using F16 as the baseline, the match rate for the first candidate of the next token was 70.0% for Q4_K_M, 22.9% for Q2_K, and 93.8% for Q8_0. The smallest quantization level to reach a 90% match rate was Q8_0, and no tested quantization reached a 95% match rate. Lowering the quantization down to Q2_K confirmed degradation, such as hitting the 1,024-token limit on the summarization task with the same line repeating 15 times.<\/p>\n<h3>Japanese Token Efficiency<\/h3>\n<div class=\"lmw-table-scroll\" tabindex=\"0\" style=\"overflow-x:auto;-webkit-overflow-scrolling:touch;max-width:100%;\">\n<table style=\"width:max-content;min-width:100%;border-collapse:collapse;\">\n<thead>\n<tr>\n<th>Tokenizer<\/th>\n<th>Tokens per 1,000 Japanese characters<\/th>\n<th>Ratio to the same text in English<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><strong>This model<\/strong><\/td>\n<td><strong>688<\/strong><\/td>\n<td><strong>1.26\u00d7<\/strong><\/td>\n<\/tr>\n<tr>\n<td>Qwen3<\/td>\n<td>688<\/td>\n<td>1.26\u00d7<\/td>\n<\/tr>\n<tr>\n<td>Llama 3.2<\/td>\n<td>744<\/td>\n<td>1.36\u00d7<\/td>\n<\/tr>\n<tr>\n<td>Gemma 3<\/td>\n<td>564<\/td>\n<td>1.03\u00d7<\/td>\n<\/tr>\n<tr>\n<td>gpt-oss<\/td>\n<td>795<\/td>\n<td>1.45\u00d7<\/td>\n<\/tr>\n<tr>\n<td>LLM-jp-3<\/td>\n<td>497<\/td>\n<td>0.85\u00d7<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p>The same Japanese text takes about as many tokens as with the Qwen3 tokenizer.<\/p>\n<p>Counted with the <code>tokenizer.json<\/code> of <a href=\"https:\/\/huggingface.co\/tencent\/WeVisDoc-4B\">tencent\/WeVisDoc-4B<\/a> on a fixed text we wrote ourselves (876 Japanese characters across news, conversation, technical docs, a formal email, travel writing and a recipe) and its English translation. Fewer tokens mean more Japanese fits in the context window.<\/p>\n<h3>Inside the GGUF File<\/h3>\n<p>File: <a href=\"https:\/\/huggingface.co\/mradermacher\/WeVisDoc-4B-GGUF\/blob\/main\/WeVisDoc-4B.Q4_K_M.gguf\">WeVisDoc-4B.Q4_K_M.gguf<\/a> (2.53GB, <code>Q4_K_M<\/code>). We read only the header (metadata) of the file, not the weights.<\/p>\n<div class=\"lmw-table-scroll\" tabindex=\"0\" style=\"overflow-x:auto;-webkit-overflow-scrolling:touch;max-width:100%;\">\n<table style=\"width:max-content;min-width:100%;border-collapse:collapse;\">\n<thead>\n<tr>\n<th>Item<\/th>\n<th>Value<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Architecture (as named in the GGUF)<\/td>\n<td><code>qwen3vl<\/code><\/td>\n<\/tr>\n<tr>\n<td>Maximum trained context length<\/td>\n<td>262,144 tokens<\/td>\n<\/tr>\n<tr>\n<td>Layers<\/td>\n<td>36<\/td>\n<\/tr>\n<tr>\n<td>Vocabulary size<\/td>\n<td>151,936<\/td>\n<\/tr>\n<tr>\n<td>Chat template<\/td>\n<td>Included (mentions tool calls)<\/td>\n<\/tr>\n<tr>\n<td>imatrix<\/td>\n<td>Not recorded in the file<\/td>\n<\/tr>\n<tr>\n<td>Weight types (share of parameters)<\/td>\n<td>Q4_K 79.9% \/ Q6_K 20.0% \/ F32 0.0%<\/td>\n<\/tr>\n<tr>\n<td>Average bits per weight<\/td>\n<td>4.91 bits<\/td>\n<\/tr>\n<tr>\n<td>Embedding \/ output layer type<\/td>\n<td>Q4_K \/ Q6_K<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p>The quant name in the file name describes the file as a whole; in practice layers mix several types. The average bits per weight is the measured file data divided by the number of weights.<\/p>\n<h3>Quality Loss by Quantization (Measured on Japanese Text)<\/h3>\n<p>We downloaded each quantization of <a href=\"https:\/\/huggingface.co\/mradermacher\/WeVisDoc-4B-GGUF\">mradermacher\/WeVisDoc-4B-GGUF<\/a> one at a time and compared it with F16 (8.22GB) on a fixed Japanese text (the opening of Natsume Soseki&#8217;s <em>Botchan<\/em>, public domain) using llama.cpp&#8217;s KL-divergence mode.<\/p>\n<div class=\"lmw-table-scroll\" tabindex=\"0\" style=\"overflow-x:auto;-webkit-overflow-scrolling:touch;max-width:100%;\">\n<p class=\"lmw-table-hint\" style=\"margin:0 0 4px;font-size:0.85em;opacity:0.7;\">\u2192 Scroll horizontally to see all columns<\/p>\n<table style=\"width:max-content;min-width:100%;border-collapse:collapse;\">\n<thead>\n<tr>\n<th>Quant<\/th>\n<th>File size<\/th>\n<th>Same top token as the baseline<\/th>\n<th>Mean KL divergence<\/th>\n<th>Perplexity vs baseline<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td><code>Q2_K<\/code><\/td>\n<td>1.67GB<\/td>\n<td>22.9%<\/td>\n<td>3.4502<\/td>\n<td>9.605\u00d7<\/td>\n<\/tr>\n<tr>\n<td><code>Q3_K_M<\/code><\/td>\n<td>2.09GB<\/td>\n<td>53.5%<\/td>\n<td>0.9025<\/td>\n<td>1.224\u00d7<\/td>\n<\/tr>\n<tr>\n<td><code>IQ4_XS<\/code><\/td>\n<td>2.32GB<\/td>\n<td>70.4%<\/td>\n<td>0.3242<\/td>\n<td>0.995\u00d7<\/td>\n<\/tr>\n<tr>\n<td><code>Q4_K_M<\/code><\/td>\n<td>2.53GB<\/td>\n<td>70.0%<\/td>\n<td>0.3195<\/td>\n<td>1.085\u00d7<\/td>\n<\/tr>\n<tr>\n<td><code>Q5_K_M<\/code><\/td>\n<td>2.94GB<\/td>\n<td>83.0%<\/td>\n<td>0.0988<\/td>\n<td>1.002\u00d7<\/td>\n<\/tr>\n<tr>\n<td><code>Q6_K<\/code><\/td>\n<td>3.38GB<\/td>\n<td>86.5%<\/td>\n<td>0.0592<\/td>\n<td>1.021\u00d7<\/td>\n<\/tr>\n<tr>\n<td><code>Q8_0<\/code><\/td>\n<td>4.37GB<\/td>\n<td>93.8%<\/td>\n<td>0.0147<\/td>\n<td>0.998\u00d7<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p>No quantization reached 95% top-token agreement.<\/p>\n<p>KL divergence closer to 0 means the next-token probabilities stay closer to the baseline. Conditions: Neoverse-N1, llama.cpp b11223, context 512 tokens, measured 2026-09-29 (JST). <em>Botchan<\/em> is a famous text the model may have seen in training, so read the numbers as a comparison between quantizations of the same model, not as an absolute score.<\/p>\n<h3>Answers by Quantization Level (Verbatim)<\/h3>\n<p>While measuring the quality loss above, we also asked the same five Japanese questions to <code>F16<\/code>, <code>Q4_K_M<\/code>, <code>Q2_K<\/code> of the same model (temperature 0, up to 1024 tokens) and publish the answers unedited, largest file first. We do not grade them.<\/p>\n<p><strong>Summarization<\/strong><\/p>\n<blockquote>\n<p>\u6b21\u306e\u8a18\u4e8b\u30923\u6587\u3067\u8981\u7d04\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n<p>\u307f\u3069\u308a\u5e02\u306f\u4eca\u5e74\u306e\u6625\u3001\u5e02\u5185\u306b\u4e09\u3064\u3042\u308b\u516c\u7acb\u56f3\u66f8\u9928\u306e\u3046\u3061\u6700\u3082\u53e4\u3044\u4e2d\u592e\u56f3\u66f8\u9928\u306e\u6539\u4fee\u3092\u7d42\u3048\u3001\u5341\u6708\u4e00\u65e5\u306b\u518d\u958b\u9928\u3057\u305f\u3002\u6539\u4fee\u306f\u7d04\u4e00\u5e74\u534a\u306b\u308f\u305f\u308a\u3001\u7dcf\u4e8b\u696d\u8cbb\u306f\u7d04\u5341\u4e8c\u5104\u5186\u3002\u3053\u306e\u3046\u3061\u7d04\u56db\u5104\u5186\u306f\u56fd\u306e\u88dc\u52a9\u91d1\u3067\u307e\u304b\u306a\u308f\u308c\u305f\u3002<\/p>\n<p>\u6700\u5927\u306e\u5909\u5316\u306f\u3001\u95b2\u89a7\u5e2d\u306e\u914d\u7f6e\u3067\u3042\u308b\u3002\u4ee5\u524d\u306f\u9928\u5185\u306e\u5ea7\u5e2d\u306e\u5927\u534a\u304c\u5b66\u7fd2\u7528\u306e\u673a\u3067\u3001\u8a66\u9a13\u671f\u9593\u306b\u306a\u308b\u3068\u4e2d\u9ad8\u751f\u3067\u57cb\u307e\u308a\u3001\u9ad8\u9f62\u306e\u5229\u7528\u8005\u3084\u5c0f\u3055\u306a\u5b50\u3069\u3082\u9023\u308c\u304c\u5ea7\u308c\u306a\u3044\u3068\u3044\u3046\u82e6\u60c5\u304c\u6bce\u5e74\u5bc4\u305b\u3089\u308c\u3066\u3044\u305f\u3002\u6539\u4fee\u5f8c\u306f\u5ea7\u5e2d\u6570\u3092\u767e\u4e8c\u5341\u5e2d\u304b\u3089\u4e8c\u767e\u5e2d\u306b\u5897\u3084\u3057\u305f\u3046\u3048\u3067\u3001\u5b66\u7fd2\u7528\u306e\u673a\u3092\u4e09\u968e\u306b\u307e\u3068\u3081\u3001\u4e00\u968e\u306f\u4f1a\u8a71\u304c\u3067\u304d\u308b\u300c\u306b\u304e\u308f\u3044\u30d5\u30ed\u30a2\u300d\u3001\u4e8c\u968e\u306f\u9759\u304b\u306b\u672c\u3092\u8aad\u3080\u300c\u95b2\u89a7\u30d5\u30ed\u30a2\u300d\u306b\u5206\u3051\u305f\u3002\u4e00\u968e\u306b\u306f\u4e88\u7d04\u5236\u306e\u30b0\u30eb\u30fc\u30d7\u5b66\u7fd2\u5ba4\u304c\u56db\u5ba4\u3068\u3001\u98f2\u307f\u7269\u3092\u6301\u3061\u8fbc\u3081\u308b\u30ab\u30d5\u30a7\u30b9\u30da\u30fc\u30b9\u3082\u8a2d\u3051\u3089\u308c\u305f\u3002<\/p>\n<p>\u8535\u66f8\u306e\u9762\u3067\u3082\u898b\u76f4\u3057\u304c\u9032\u3093\u3060\u3002\u8cb8\u51fa\u306e\u5c11\u306a\u3044\u53e4\u3044\u8cc7\u6599\u7d04\u4e09\u4e07\u518a\u3092\u5e02\u5185\u306e\u4ed6\u306e\u9928\u3084\u4fdd\u5b58\u66f8\u5eab\u306b\u79fb\u3057\u3001\u4ee3\u308f\u308a\u306b\u5b50\u3069\u3082\u5411\u3051\u306e\u7d75\u672c\u3068\u3001\u4ecb\u8b77\u30fb\u5b50\u80b2\u3066\u30fb\u4ed5\u4e8b\u63a2\u3057\u3068\u3044\u3063\u305f\u66ae\u3089\u3057\u306b\u95a2\u308f\u308b\u5b9f\u7528\u66f8\u3092\u7d04\u4e00\u4e07\u4e94\u5343\u518a\u5897\u3084\u3057\u305f\u3002\u96fb\u5b50\u66f8\u7c4d\u306e\u8cb8\u51fa\u30b5\u30fc\u30d3\u30b9\u3082\u59cb\u307e\u308a\u3001\u5229\u7528\u767b\u9332\u3092\u3059\u308c\u3070\u81ea\u5b85\u304b\u3089\u4e00\u5ea6\u306b\u4e09\u518a\u307e\u3067\u3001\u4e8c\u9031\u9593\u501f\u308a\u3089\u308c\u308b\u3002<\/p>\n<p>\u5efa\u7269\u305d\u306e\u3082\u306e\u306b\u3082\u624b\u304c\u5165\u3063\u305f\u3002\u7bc9\u56db\u5341\u5e74\u3092\u8d85\u3048\u3066\u3044\u305f\u7a7a\u8abf\u8a2d\u5099\u3092\u5165\u308c\u66ff\u3048\u3001\u5c4b\u4e0a\u306b\u306f\u592a\u967d\u5149\u30d1\u30cd\u30eb\u3092\u8a2d\u7f6e\u3057\u3066\u3001\u9928\u5185\u3067\u4f7f\u3046\u96fb\u529b\u306e\u7d04\u4e8c\u5272\u3092\u307e\u304b\u306a\u3048\u308b\u3088\u3046\u306b\u3057\u305f\u3002\u8eca\u3044\u3059\u306e\u5229\u7528\u8005\u304b\u3089\u8981\u671b\u306e\u591a\u304b\u3063\u305f\u30a8\u30ec\u30d9\u30fc\u30bf\u30fc\u306f\u4e00\u57fa\u304b\u3089\u4e8c\u57fa\u306b\u5897\u3048\u3001\u66f8\u67b6\u306e\u9593\u306e\u901a\u8def\u3082\u4ee5\u524d\u3088\u308a\u4e09\u5341\u30bb\u30f3\u30c1\u5e83\u3052\u3089\u308c\u305f\u3002\u5165\u53e3\u306b\u306f\u6bb5\u5dee\u306e\u306a\u3044\u30b9\u30ed\u30fc\u30d7\u304c\u65b0\u305f\u306b\u8a2d\u3051\u3089\u308c\u3001\u6388\u4e73\u5ba4\u3068\u591a\u76ee\u7684\u30c8\u30a4\u30ec\u3082\u5404\u968e\u306b\u4e00\u3064\u305a\u3064\u7f6e\u304b\u308c\u305f\u3002<\/p>\n<p>\u4e00\u65b9\u3067\u8ab2\u984c\u3082\u6b8b\u308b\u3002\u99d0\u8eca\u5834\u306f\u6539\u4fee\u524d\u3068\u540c\u3058\u56db\u5341\u53f0\u306e\u307e\u307e\u3067\u3001\u9031\u672b\u306b\u306f\u8fd1\u304f\u306e\u9053\u8def\u306b\u9806\u756a\u5f85\u3061\u306e\u8eca\u304c\u4e26\u3093\u3060\u3002\u5e02\u306f\u96a3\u63a5\u3059\u308b\u5e02\u6709\u5730\u3092\u81e8\u6642\u99d0\u8eca\u5834\u3068\u3057\u3066\u958b\u653e\u3057\u305f\u304c\u3001\u6052\u4e45\u7684\u306a\u5bfe\u7b56\u306f\u6c7a\u307e\u3063\u3066\u3044\u306a\u3044\u3002\u307e\u305f\u3001\u958b\u9928\u6642\u9593\u3092\u5e73\u65e5\u306e\u591c\u4e5d\u6642\u307e\u3067\u5ef6\u3070\u3057\u305f\u3053\u3068\u3067\u8077\u54e1\u306e\u8ca0\u62c5\u304c\u5897\u3048\u3066\u304a\u308a\u3001\u8077\u54e1\u56e3\u4f53\u306f\u4eba\u54e1\u306e\u8ffd\u52a0\u3092\u6c42\u3081\u3066\u3044\u308b\u3002<\/p>\n<p>\u5229\u7528\u8005\u306e\u53d7\u3051\u6b62\u3081\u306f\u304a\u304a\u3080\u306d\u597d\u610f\u7684\u3060\u3002\u9031\u306b\u4e8c\u5ea6\u3001\u5b6b\u3092\u9023\u308c\u3066\u6765\u308b\u3068\u3044\u3046\u4e03\u5341\u4ee3\u306e\u5973\u6027\u306f\u300c\u524d\u306f\u5b50\u3069\u3082\u304c\u58f0\u3092\u51fa\u3059\u3068\u5468\u308a\u306e\u76ee\u304c\u6c17\u306b\u306a\u3063\u305f\u304c\u3001\u4e00\u968e\u306a\u3089\u6c17\u517c\u306d\u306a\u304f\u7d75\u672c\u3092\u8aad\u3093\u3067\u3042\u3052\u3089\u308c\u308b\u300d\u3068\u8a71\u3059\u3002\u4e00\u65b9\u3001\u53d7\u9a13\u3092\u63a7\u3048\u305f\u9ad8\u6821\u751f\u306e\u4e2d\u306b\u306f\u300c\u5b66\u7fd2\u5e2d\u304c\u4e09\u968e\u3060\u3051\u306b\u306a\u308a\u3001\u5915\u65b9\u306f\u3059\u3050\u306b\u6e80\u5e2d\u306b\u306a\u308b\u300d\u3068\u4e0d\u6e80\u3092\u6f0f\u3089\u3059\u4eba\u3082\u3044\u308b\u3002\u56f3\u66f8\u9928\u306f\u3001\u5b66\u7fd2\u5e2d\u306e\u7a7a\u304d\u72b6\u6cc1\u3092\u30db\u30fc\u30e0\u30da\u30fc\u30b8\u3067\u78ba\u8a8d\u3067\u304d\u308b\u4ed5\u7d44\u307f\u3092\u5e74\u5185\u306b\u5c0e\u5165\u3059\u308b\u4e88\u5b9a\u3060\u3002<\/p>\n<p>\u518d\u958b\u9928\u306b\u5408\u308f\u305b\u3066\u3001\u50ac\u3057\u3082\u5897\u3048\u305f\u3002\u6bce\u9031\u571f\u66dc\u306e\u5348\u524d\u306b\u306f\u8aad\u307f\u805e\u304b\u305b\u4f1a\u3001\u6708\u306b\u4e00\u5ea6\u306f\u5730\u5143\u306e\u8fb2\u5bb6\u3084\u5546\u5e97\u4e3b\u3092\u8b1b\u5e2b\u306b\u62db\u304f\u300c\u307e\u3061\u306e\u5148\u751f\u300d\u8b1b\u5ea7\u304c\u958b\u304b\u308c\u308b\u3002\u5341\u4e00\u6708\u306b\u306f\u3001\u5e02\u5185\u306e\u4e2d\u5b66\u751f\u304c\u9078\u3093\u3060\u672c\u3092\u7d39\u4ecb\u3059\u308b\u5c55\u793a\u3082\u4e88\u5b9a\u3055\u308c\u3066\u3044\u308b\u3002<\/p>\n<p>\u9928\u9577\u306e\u5c71\u672c\u548c\u5b50\u3055\u3093\u306f\u300c\u672c\u3092\u501f\u308a\u308b\u5834\u6240\u304b\u3089\u3001\u4eba\u304c\u96c6\u307e\u3063\u3066\u904e\u3054\u3059\u5834\u6240\u3078\u5909\u3048\u305f\u304b\u3063\u305f\u3002\u518d\u958b\u9928\u304b\u3089\u306e\u4e00\u304b\u6708\u3067\u6765\u9928\u8005\u306f\u4ee5\u524d\u306e\u4e00\u30fb\u516b\u500d\u306b\u306a\u3063\u305f\u300d\u3068\u8a71\u3059\u3002\u5e02\u306f\u6765\u5e74\u5ea6\u3001\u6b8b\u308b\u4e8c\u3064\u306e\u56f3\u66f8\u9928\u306b\u3064\u3044\u3066\u3082\u5229\u7528\u8005\u3078\u306e\u805e\u304d\u53d6\u308a\u3092\u884c\u3044\u3001\u6539\u4fee\u306e\u5fc5\u8981\u6027\u3092\u691c\u8a0e\u3059\u308b\u65b9\u91dd\u3060\u3002<\/p>\n<\/blockquote>\n<p><em>(Summarize the following article in three sentences. (The article, about 1,200 Japanese characters, describes the renovation and reopening of a city&#8217;s central library: the budget, the new floor layout and seating, changes to the collection and e-book lending, building and accessibility work, remaining issues such as parking and staff workload, users&#8217; reactions, new events, and the city&#8217;s next steps.))<\/em><\/p>\n<p><code>F16<\/code> (baseline) 8.22GB<\/p>\n<pre><code class=\"language-text\">\u307f\u3069\u308a\u5e02\u306f\u6625\u306b\u4e2d\u592e\u56f3\u66f8\u9928\u3092\u6539\u4fee\u3057\u3001\u5341\u6708\u4e00\u65e5\u306b\u518d\u958b\u9928\u3057\u305f\u3002\u7d04\u4e00\u5e74\u534a\u306b\u308f\u305f\u308b\u5de5\u4e8b\u306f\u7dcf\u4e8b\u696d\u8cbb\u7d04\u5341\u4e8c\u5104\u5186\u3067\u3001\u7d04\u56db\u5104\u5186\u306f\u56fd\u306e\u88dc\u52a9\u91d1\u3067\u8cc4\u308f\u308c\u305f\u3002\n\n\u6700\u5927\u306e\u5909\u5316\u306f\u5ea7\u5e2d\u914d\u7f6e\u3067\u3001\u5b66\u7fd2\u7528\u306e\u673a\u304c\u4e3b\u3060\u3063\u305f\u5ea7\u5e2d\u3092\u9ad8\u9f62\u8005\u3084\u5b50\u3069\u3082\u9023\u308c\u306b\u914d\u616e\u3057\u305f\u300c\u306b\u304e\u308f\u3044\u30d5\u30ed\u30a2\u300d\u3084\u300c\u95b2\u89a7\u30d5\u30ed\u30a2\u300d\u306b\u5206\u3051\u3001\u96fb\u5b50\u66f8\u7c4d\u30b5\u30fc\u30d3\u30b9\u3082\u59cb\u307e\u308a\u3001\u5efa\u7269\u306b\u3082\u7a7a\u8abf\u3084\u592a\u967d\u5149\u30d1\u30cd\u30eb\u3001\u30a8\u30ec\u30d9\u30fc\u30bf\u30fc\u5897\u8a2d\u306a\u3069\u624b\u304c\u5165\u308c\u3089\u308c\u305f\u3002\n\n\u8ab2\u984c\u306f\u99d0\u8eca\u5834\u306e\u4e0d\u8db3\u3084\u8077\u54e1\u8ca0\u62c5\u3060\u304c\u3001\u5229\u7528\u8005\u306e\u53d7\u3051\u5165\u308c\u306f\u304a\u304a\u3080\u306d\u597d\u610f\u7684\u3067\u3001\u50ac\u3057\u3082\u5897\u3048\u3001\u5e02\u306f\u6765\u5e74\u5ea6\u3001\u6b8b\u308b\u56f3\u66f8\u9928\u306e\u6539\u4fee\u691c\u8a0e\u65b9\u91dd\u3092\u793a\u3059\u3002\n<\/code><\/pre>\n<p><code>Q4_K_M<\/code> 2.53GB<\/p>\n<pre><code class=\"language-text\">\u307f\u3069\u308a\u5e02\u306f\u6625\u306b\u4e2d\u592e\u56f3\u66f8\u9928\u3092\u6539\u4fee\u3057\u3001\u5341\u6708\u4e00\u65e5\u306b\u518d\u958b\u9928\u3057\u305f\u3002\u7d04\u4e00\u5e74\u534a\u306e\u5de5\u671f\u3068\u7d04\u5341\u4e8c\u5104\u5186\u306e\u4e8b\u696d\u8cbb\uff08\u7d04\u56db\u5104\u5186\u304c\u56fd\u306e\u88dc\u52a9\u91d1\uff09\u3067\u3001\u4e3b\u306a\u5909\u5316\u306f\u5ea7\u5e2d\u914d\u7f6e\u3084\u5efa\u7269\u306e\u8a2d\u5099\u6539\u5584\u306a\u3069\u3060\u3002\n\n\u6539\u4fee\u3067\u306f\u5b66\u7fd2\u5e2d\u3092\u4e09\u968e\u306b\u307e\u3068\u3081\u3001\u4e00\u968e\u3092\u4f1a\u8a71\u30d5\u30ed\u30a2\u3084\u95b2\u89a7\u30d5\u30ed\u30a2\u306b\u5206\u3051\u3001\u96fb\u5b50\u66f8\u7c4d\u30b5\u30fc\u30d3\u30b9\u3084\u8eca\u3044\u3059\u5bfe\u7b56\u3001\u7a7a\u8abf\u3084\u592a\u967d\u5149\u30d1\u30cd\u30eb\u8a2d\u7f6e\u306a\u3069\u3001\u5229\u7528\u8005\u3084\u74b0\u5883\u3078\u306e\u914d\u616e\u304c\u9032\u307f\u3001\u99d0\u8eca\u5834\u3084\u8077\u54e1\u8ca0\u62c5\u306a\u3069\u306e\u8ab2\u984c\u3082\u6b8b\u3063\u3066\u3044\u308b\u3002\n\n\u5229\u7528\u8005\u306f\u6574\u4f53\u7684\u306b\u597d\u610f\u7684\u3067\u3001\u9ad8\u9f62\u8005\u3084\u5bb6\u65cf\u9023\u308c\u306e\u6765\u9928\u5897\u52a0\u3092\u559c\u3076\u4e00\u65b9\u3001\u53d7\u9a13\u751f\u3084\u5b66\u751f\u306e\u4e0d\u6e80\u3082\u3042\u308a\u3001\u56f3\u66f8\u9928\u306f\u7a7a\u304d\u72b6\u6cc1\u306e\u78ba\u8a8d\u4ed5\u7d44\u307f\u3092\u5e74\u5185\u306b\u5c0e\u5165\u4e88\u5b9a\u3002\u5e02\u306f\u6765\u5e74\u5ea6\u306f\u6b8b\u308b\u4e8c\u3064\u306e\u56f3\u66f8\u9928\u306b\u3064\u3044\u3066\u3082\u5229\u7528\u8005\u8abf\u67fb\u3092\u884c\u3044\u3001\u6539\u4fee\u306e\u5fc5\u8981\u6027\u3092\u691c\u8a0e\u3059\u308b\u65b9\u91dd\u3060\u3002\n<\/code><\/pre>\n<p><code>Q2_K<\/code> 1.67GB<\/p>\n<pre><code class=\"language-text\">\u307fdo\u0161i: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u6700\u53e4\u3001\u4e2d\u592e\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u6700\u53e4\u3001\u4e2d\u592e\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u6700\u53e4\u3001\u4e2d\u592e\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u6700\u53e4\u3001\u4e2d\u592e\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539\u4e09\u6539\u3001\u6539\u4fee\u3001\u6539\u4fee\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u6700\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u6700\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u6700\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u6700\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u6700\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\u56f3\u66f8\u9928\u306e\u5185\u3001\u66f4\u53e4\u3001\u4e2d\u592e\u3001\u6539 \u4f5c\n\u5efa\u7269: \u5e02\u5185\u306f\u3001\u4e09\u3064\n<\/code><\/pre>\n<p><em>(Cut off at the 1024-token limit.)<\/em><\/p>\n<p><strong>Formal business email<\/strong><\/p>\n<blockquote>\n<p>\u53d6\u5f15\u5148\u306e\u62c5\u5f53\u8005\u306b\u3001\u6765\u9031\u706b\u66dc\u65e5\u306e\u6253\u3061\u5408\u308f\u305b\u3092\u6c34\u66dc\u65e5\u306b\u5909\u66f4\u3057\u3066\u307b\u3057\u3044\u3068\u304a\u9858\u3044\u3059\u308b\u30e1\u30fc\u30eb\u3092\u3001\u4e01\u5be7\u306a\u656c\u8a9e\u3067\u66f8\u3044\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n<\/blockquote>\n<p><em>(Write an email in polite Japanese (keigo) asking a client&#8217;s contact person to move next Tuesday&#8217;s meeting to Wednesday.)<\/em><\/p>\n<p><code>F16<\/code> (baseline) 8.22GB<\/p>\n<pre><code class=\"language-text\">\u4ef6\u540d\uff1a\u6765\u9031\u706b\u66dc\u65e5\u306e\u6253\u3061\u5408\u308f\u305b\u3092\u6c34\u66dc\u65e5\u306b\u5909\u66f4\u3059\u308b\u304a\u9858\u3044\n\n\u3007\u3007\u69d8\n\n\u8cb4\u793e\u306e\u53d6\u5f15\u5148\u3068\u3057\u3066\u3001\u3044\u3064\u3082\u304a\u4e16\u8a71\u306b\u306a\u3063\u3066\u304a\u308a\u307e\u3059\u3002\u3053\u306e\u5ea6\u306f\u3001\u6765\u9031\u706b\u66dc\u65e5\u306e\u6253\u3061\u5408\u308f\u305b\u6642\u9593\u3092\u6c34\u66dc\u65e5\u306b\u5909\u66f4\u3057\u3066\u9802\u6234\u3067\u304d\u308c\u3070\u3068\u5b58\u3058\u307e\u3059\u3002\u3054\u90fd\u5408\u304c\u4f3a\u3048\u308b\u304b\u3001\u304a\u77e5\u3089\u305b\u8cdc\u308a\u307e\u3059\u3088\u3046\u304a\u9858\u3044\u7533\u3057\u4e0a\u3052\u307e\u3059\u3002\n\n\u4e07\u304c\u4e00\u90fd\u5408\u304c\u3064\u304b\u305a\u3001\u3054\u90fd\u5408\u306b\u6cbf\u3048\u306a\u3044\u5834\u5408\u306f\u3001\u3054\u9023\u7d61\u8cdc\u308a\u307e\u3059\u3088\u3046\u304a\u9858\u3044\u7533\u3057\u4e0a\u3052\u307e\u3059\u3002\u4f55\u5352\u3001\u3088\u308d\u3057\u304f\u304a\u9858\u3044\u3044\u305f\u3057\u307e\u3059\u3002\n\n\u656c\u5177\n\n\u25cb\u25cb\u25cb \u4ece\n<\/code><\/pre>\n<p><code>Q4_K_M<\/code> 2.53GB<\/p>\n<pre><code class=\"language-text\">\u4ef6\u540d\uff1a\u6765\u9031\u706b\u66dc\u65e5\u6253\u3061\u5408\u308f\u305b\u306e\u65e5\u6642\u5909\u66f4\u306e\u304a\u9858\u3044\n\n\u3007\u3007\u69d8\n\n\u8cb4\u793e\u306e\u53d6\u5f15\u5148\u3068\u3057\u3066\u3001\u3044\u3064\u3082\u304a\u4e16\u8a71\u306b\u306a\u3063\u3066\u304a\u308a\u307e\u3059\u3002\u672c\u4ef6\u306b\u95a2\u3057\u307e\u3057\u3066\u306f\u3001\u6765\u9031\u706b\u66dc\u65e5\u306e\u6253\u3061\u5408\u308f\u305b\u3092\u6c34\u66dc\u65e5\u306b\u5909\u66f4\u3057\u3066\u9802\u6234\u4e86\u51fa\u6765\u4e2a\u5de5\u4f5c\u65e5\u3092\u671b\u307f\u307e\u3059\u3002\n\n\u4e0a\u8a18\u306e\u5909\u66f4\u3092\u5e0c\u671b\u3044\u305f\u3057\u307e\u3059\u306e\u3067\u3001\u3054\u4e86\u627f\u4e0b\u3055\u3044\u307e\u3059\u3088\u3046\u304a\u9858\u3044\u7533\u3057\u4e0a\u3052\u307e\u3059\u3002\n\n\u4f55\u5352\u3088\u308d\u3057\u304f\u304a\u9858\u3044\u3044\u305f\u3057\u307e\u3059\u3002\n\n\u25cb\u25cb\u69d8\n<\/code><\/pre>\n<p><code>Q2_K<\/code> 1.67GB<\/p>\n<pre><code class=\"language-text\">\u53d6\u5f15\u5148\u306e\u62c5\u5f53\u8005\u3055\u307e\u3001\u6765\u9031\u706b\u66dc\u65e5\u306e\u53d6\u308a\u6242\u308a\u3092\u5909\u66f4\u3057\u3066\u307b\u3057\u3044\u3068\u5b58\u3058andum\u3092\u3001\u4e01\u8f9e\u306a\u656c\u8a9e\u3067\u304a\u9858\u3044\u3057\u307e\u3059\u3002\n<\/code><\/pre>\n<p><strong>Word problem<\/strong><\/p>\n<blockquote>\n<p>\u308a\u3093\u3054\u304c1\u500b120\u5186\u3001\u307f\u304b\u3093\u304c1\u500b45\u5186\u3067\u3059\u3002\u308a\u3093\u3054\u30923\u500b\u3068\u307f\u304b\u3093\u30928\u500b\u8cb7\u3063\u3066\u30011,000\u5186\u672d\u3067\u6255\u3044\u307e\u3057\u305f\u3002\u304a\u3064\u308a\u306f\u3044\u304f\u3089\u3067\u3059\u304b\u3002\u8003\u3048\u65b9\u3082\u793a\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n<\/blockquote>\n<p><em>(Apples cost 120 yen each and mandarins cost 45 yen each. You buy 3 apples and 8 mandarins and pay with a 1,000-yen bill. How much change do you get? Show your reasoning.)<\/em><\/p>\n<p><code>F16<\/code> (baseline) 8.22GB<\/p>\n<pre><code class=\"language-text\">\u308a\u3093\u3054\u304c1\u500b120\u5186\u3001\u307f\u304b\u3093\u304c1\u500b45\u5186\u3067\u3059\u3002\n\n\u308a\u3093\u3054\u30923\u500b\u3068\u307f\u304b\u3093\u30928\u500b\u8cb7\u3063\u305f\u306e\u3067\u3001\n\n\u308a\u3093\u3054\u306e\u6599\u91d1\uff1a120\u5186 \u00d7 3\u500b = 360\u5186\n\n\u307f\u304b\u3093\u306e\u6599\u91d1\uff1a45\u5186 \u00d7 8\u500b = 360\u5186\n\n\u5408\u8a08\u306e\u6599\u91d1\uff1a360\u5186 + 360\u5186 = 720\u5186\n\n1,000\u5186\u672d\u3067\u6255\u3063\u305f\u306e\u3067\u3001\n\n\u304a\u3064\u308a\uff1a1,000\u5186 - 720\u5186 = 280\u5186\n\n**\u7b54\u3048\uff1a280\u5186**\n\n**\u8003\u3048\u65b9\uff1a**\n1. \u5404\u679c\u7269\u306e\u5358\u4fa1\u3068\u6570\u91cf\u3092\u304b\u3051\u7b97\u3057\u3066\u3001\u305d\u308c\u305e\u308c\u306e\u6599\u91d1\u3092\u8a08\u7b97\u3002\n2. \u4e21\u8005\u306e\u6599\u91d1\u3092\u8db3\u3057\u7b97\u3057\u3066\u3001\u5408\u8a08\u6599\u91d1\u3092\u6c42\u3081\u308b\u3002\n3. \u652f\u6255\u3063\u305f\u91d1\u984d\uff081,000\u5186\uff09\u304b\u3089\u5408\u8a08\u6599\u91d1\u3092\u5f15\u304f\u3068\u3001\u304a\u3064\u308a\u306e\u984d\u304c\u5f97\u3089\u308c\u308b\u3002\n<\/code><\/pre>\n<p><code>Q4_K_M<\/code> 2.53GB<\/p>\n<pre><code class=\"language-text\">\u308a\u3093\u3054\u304c1\u500b120\u5186\u3001\u307f\u304b\u3093\u304c1\u500b45\u5186\u3067\u3059\u3002\n\n\u308a\u3093\u3054\u30923\u500b\u3068\u307f\u304b\u3093\u30928\u500b\u8cb7\u3063\u305f\u306e\u3067\u3001\n\n\u308a\u3093\u3054\u306e\u5408\u8a08\u91d1\u984d\uff1a120\u5186 \u00d7 3 = 360\u5186\n\n\u307f\u304b\u3093\u306e\u5408\u8a08\u91d1\u984d\uff1a45\u5186 \u00d7 8 = 360\u5186\n\n\u5408\u8a08\u91d1\u984d\uff1a360\u5186 + 360\u5186 = 720\u5186\n\n1,000\u5186\u672d\u3067\u6255\u3063\u305f\u306e\u3067\u3001\n\n\u304a\u3064\u308a\uff1a1,000\u5186 - 720\u5186 = 280\u5186\n\n\u3088\u3063\u3066\u3001\u304a\u3064\u308a\u306f**280\u5186**\u3067\u3059\u3002\n\n### \u8003\u3048\u65b9\u306e\u307e\u3068\u3081\uff1a\n\n1. \u5404\u7a2e\u679c\u7269\u306e\u500b\u6570\u3068\u5358\u4fa1\u3092 multiply\u3057\u3066\u5408\u8a08\u91d1\u984d\u3092\u8a08\u7b97\u3002\n2. \u4e21\u8005\u306e\u5408\u8a08\u91d1\u984d\u3092\u8db3\u3057\u7b97\u3057\u3066\u5168\u4f53\u306e\u5408\u8a08\u984d\u3092\u6c42\u3081\u308b\u3002\n3. \u652f\u6255\u3063\u305f\u91d1\u984d\uff081,000\u5186\uff09\u304b\u3089\u5408\u8a08\u984d\u3092\u5f15\u304f\u3068\u304a\u3064\u308a\u306e\u984d\u304c\u6c42\u307e\u308b\u3002\n\n\u8a08\u7b97\u5f0f\uff1a\n\n$$\n1,000 - (120 \\times 3 + 45 \\times 8) = 1,000 - (360 + 360) = 1,000 - 720 = 280\n$$\n\n\u7b54\u3048\uff1a**280\u5186**\n<\/code><\/pre>\n<p><code>Q2_K<\/code> 1.67GB<\/p>\n<pre><code class=\"language-text\">\u89e3\u304d\u65b9\u7a0b.\n\n(1) 1,000 yen (100 yen \u6279) (2) 1,000 yen (100 yen \u6279) (3) 1,000 yen (100 yen \u6279) (4) 1,000 yen (100 yen \u6279) (5) 1,000 \u00a5\n\n(1) 1,000 yen (100 \u00a5) (2) 1,000 \u00a5 (100 \u00a5) (3) 1,001 \u00a5 (100 \u00a5) (4) 1,001 \u00a5 (100 \u00a5) (5) 1,001 \u00a5 (100 \u00a5) (6) 1,001 \u00a5 (100 \u00a5) (7) 1,001 \u00a5 (100 \u00a5) (8) 1,001 \u00a5 (100 \u00a5) (9) 1,001 \u00a5 (100 \u00a5) (10) 1,001 \u00a5 (100 \u00a5) (11) 1,001 \u00a5 (100 \u00a5) (12) 1,001 \u00a5 (100 \u00a5) (13) 1,001 \u00a5 (100 \u00a5) (14) 1,001 \u00a5 (100 \u00a5) (15) 1,001 \u00a5 (100 \u00a5) (16) 1,001 \u00a5 (100 \u00a5) (17) 1,001 \u00a5 (100 \u00a5) (18) 1,001 \u00a5 (100 \u00a5) (19)1 1,001 \u00a5 (100 \u00a5) (20) 1,001 \u00a5 (100 \u00a5) (21) 1,001 \u00a5 (100 \u00a5) (22) 1,001 \u00a5 (100 \u00a5) (23) 1,001 \u00a5 (100 \u00a5) (24) 1,001 \u00a5 (100 \u00a5) (25) 1,001 \u00a5 (100 \u00a5) (26) 1,001 \u00a5 (100 \u00a5) (27) 1,001 \u00a5 (100 \u00a5) (28) 1,001 \u00a5 (100 \u00a5) (29)1 1,011 \u00a5 (101 \u00a5) (30) 1,011 \u00a5 (101 \u00a5) (31) 1,011 \u00a5 (101 \u00a5) (32) 1,011 \u00a5 (101 \u00a5) (33) 1,011 \u00a5 (101 \u00a5) (34) 1,011 \u00a5 (101 \u00a5) (35) 1,011 \u00a5 (101 \u00a5) (36) 1,011 \u00a5 (101 \u00a5) (37) 1,011 \u00a5 (101 \u00a5) (38) 1,011 \u00a5 (101 \u00a5) (39)1 1,011 \u00a5 (101) (40) 1,011 \u00a5 (101) (41) 1,011 \u00a5 (101) (42) 1,011 \u00a5 (101) (43) 1,011 \u00a5 (101) (44) 1,011 \u00a5 (101) (45) 1,011 \u00a5 (101) (46) 1,011 \u00a5 (101) (47) 1,011 \u00a5 (101) (48) 1,011 \u00a5 (101) (49) 1,011 \u00a5 (101) (50) 1,011 \u00a5 (101) (50) 1,011 \u00a5 (101) (51) 1,011 \u00a5 (101) (52) 1,011 \u00a5 (101) (53) 1,011 \u00a5 (101) (54) 1,011 \u00a5 (101) (55) 1,011 \u00a5 (101) (56) 1,0\n<\/code><\/pre>\n<p><em>(Cut off at the 1024-token limit.)<\/em><\/p>\n<p>Reference: The correct answer is 280 yen (120\u00d73 = 360, 45\u00d78 = 360, 1,000 \u2212 720 = 280).<\/p>\n<p><strong>Knowledge<\/strong><\/p>\n<blockquote>\n<p>\u65e5\u672c\u306e\u90fd\u9053\u5e9c\u770c\u306e\u3046\u3061\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3092\u3059\u3079\u3066\u6319\u3052\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n<\/blockquote>\n<p><em>(List all of Japan&#8217;s prefectures that do not border the sea.)<\/em><\/p>\n<p><code>F16<\/code> (baseline) 8.22GB<\/p>\n<pre><code class=\"language-text\">\u65e5\u672c\u306e\u90fd\u9053\u5e9c\u770c\u306e\u3046\u3061\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\uff08\u300c\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u90fd\u9053\u5e9c\u770c\u300d\uff09\u3092\u6319\u3052\u307e\u3059\u3002\u6ce8\u610f\u70b9\u3068\u3057\u3066\u306f\uff1a\n\n- \u300c\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u300d\uff1d\u300c\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u300d\uff1d\u300c\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u5730\u65b9\u81ea\u6cbb\u4f53\u300d\n- \u4f8b\u5916\u3068\u3057\u3066\u3001\u4e00\u90e8\u306e\u770c\u304c\u300c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u304c\u3001\u884c\u653f\u7684\u306b\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u300d\u3068\u3055\u308c\u308b\u5834\u5408\u3082\u3042\u308a\u307e\u3059\u304c\u3001\u3053\u3053\u3067\u306f\u4e00\u822c\u7684\u306a\u5b9a\u7fa9\u3067\u6574\u7406\u3057\u307e\u3059\u3002\n- \u7fa4\u99ac\u770c\u3001\u5c71\u68a8\u770c\u3001\u9577\u91ce\u770c\u3001\u65b0\u6f5f\u770c\u3001\u5bcc\u5c71\u770c\u3001\u77f3\u5ddd\u770c\u3001\u798f\u4e95\u770c\u3001\u5c71\u53e3\u770c\u3001\u9999\u5ddd\u770c\u3001\u5fb3\u5cf6\u770c\u3001\u9ad8\u77e5\u770c\u3001\u611b\u5a9b\u770c\u3001\u56db\u56fd\u5730\u65b9\u306e\u4e00\u90e8\u770c\u306a\u3069\u306f\u300c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u300d\u3053\u3068\u304c\u591a\u3044\u3067\u3059\u3002\n- \u4e00\u65b9\u3067\u3001\u5317\u6d77\u9053\u306e\u4e00\u90e8\u770c\uff08\u5317\u6d77\u9053\u30fb\u6839\u5ba4\u5e73\u91ce\u30fb\u6839\u5ba4\u753a\u30fb\u6839\u5ba4\u6751\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751\uff09\u306f\u300c\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u300d\u3053\u3068\u304c\u4e00\u822c\u7684\u3067\u3059\u3002\n\n## \u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u90fd\u9053\u5e9c\u770c\uff08\u6b63\u5f0f\u306a\u540d\u79f0\uff09\n\n1. **\u7fa4\u99ac\u770c**\n\n2. **\u5c71\u68a8\u770c**\n\n3. **\u9577\u91ce\u770c**\n\n4. **\u65b0\u6f5f\u770c**\n\n5. **\u5bcc\u5c71\u770c**\n\n6. **\u77f3\u5ddd\u770c**\n\n7. **\u798f\u4e95\u770c**\n\n8. **\u5c71\u53e3\u770c**\n\n9. **\u9999\u5ddd\u770c**\n\n10. **\u5fb3\u5cf6\u770c**\n\n11. **\u9ad8\u77e5\u770c**\n\n12. **\u611b\u5a9b\u770c**\n\n13. **\u5c71\u68a8\u770c**\uff08\u4e0a\u8a18\u3068\u91cd\u8907\uff09\n\n14. **\u65b0\u6f5f\u770c**\uff08\u4e0a\u8a18\u3068\u91cd\u8907\uff09\n\n15. **\u5317\u6d77\u9053\u30fb\u6839\u5ba4\u5e73\u91ce\u30fb\u6839\u5ba4\u753a\u30fb\u6839\u5ba4\u6751\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n16. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n17. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n18. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n19. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n20. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n21. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n22. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n23. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n24. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n25. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n26. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n27. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n28. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n29. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n30. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n31. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n32. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n33. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n34. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n35. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n36. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n37. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n38. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n39. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n40. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n41. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n42. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n43. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n44. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n45. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\u753a\u30fb\u659c\u91cc\u6751**\uff08\u4e00\u90e8\u306e\u81ea\u6cbb\u4f53\uff09\n\n46. **\u5317\u6d77\u9053\u30fb\u659c\u91cc\n<\/code><\/pre>\n<p><em>(Cut off at the 1024-token limit.)<\/em><\/p>\n<p><code>Q4_K_M<\/code> 2.53GB<\/p>\n<pre><code class=\"language-text\">\u65e5\u672c\u306e\u90fd\u9053\u5e9c\u770c\u306e\u3046\u3061\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3092\u6319\u3052\u307e\u3059\u3002\u65e5\u672c\u5217\u5cf6\u306f\u3001\u6d77\u306b\u9762\u3059\u308b\u770c\u304c\u591a\u3044\u3067\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u305b\u3093\u3002\u5177\u4f53\u7684\u306b\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3067\u3059\uff1a\n\n### **\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u90fd\u9053\u5e9c\u770c\uff08\u770c\uff09**\n\n1. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n   - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n   - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n2. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n   - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n   - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n3. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n   - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n   - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n4. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n   - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n   - \u305f\u3060\u3057\u3001\u5317\u6d77\u3002\n5. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n   - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n   - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n6. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n   - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n   - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n7. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n   - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n   - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n8. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n   - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n   - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n9. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n   - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n   - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n10. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n    - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n    - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n11. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n    - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n    - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n12. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n    - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n    - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n13. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n    - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n    - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n14. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n    - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n    - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n15. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n    - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n    - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u3068\u3057\u3066\u9664\u5916\u3055\u308c\u307e\u3059\u3002\n\n16. **\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09**\n    - \u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u304c\u3001\u4e00\u90e8\u306e\u770c\uff08\u5317\u6d77\u9053\uff09\u306f\u6d77\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n    - \u305f\u3060\u3057\u3001\u5317\u6d77\u9053\u5168\u57df\u304c\u6d77\u306b\u9762\u3057\u3066\u3044\u308b\u305f\u3081\u3001\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\n<\/code><\/pre>\n<p><em>(Cut off at the 1024-token limit.)<\/em><\/p>\n<p><code>Q2_K<\/code> 1.67GB<\/p>\n<pre><code class=\"language-text\">\u65e5\u672c\u5168\u56fd\u5404\u5730\u306e\u3046\u3061\u3001\u6d77\u306b\u9762\u3057\u306a\u3044\u770c\u3092\u3059\u3079\u3066\u6319\u3052\u307e\u3059\u3002\n\n1. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u65e5\u672c\u6d77\uff08\u65e5\u672c\u6d77\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n2. \u5317\u6d77aut\uff08\u5317\u6d77aut\uff09\uff1a\u5317\u6d77aut\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77aut\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n3. \u5317\u6d77\u6c49\uff08\u5317\u6d77\u6c49\uff09\uff1a\u5317\u6d77\u6c49\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77\u6c49\uff08\u5317\u6d77\u6c49\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n4. \u5317\u6d77\u5c9b\uff08\u5317\u6d77\u5c9b\uff09\uff1a\u5317\u6d77\u5c9b\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u6d77\u5dba\uff08\u65e5\u672c\u6d77\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n5. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u65e5\u672c\u6d77\uff08\u65e5\u672c\u6d77\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n6. \u5317\u6d77aut\uff08\u5317\u6d77aut\uff09\uff1a\u5317\u6d77aut\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77aut\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n7. \u5317\u6d77\u6c49\uff08\u5317\u6d77\u6c49\uff09\uff1a\u5317\u6d77aut\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77aut\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n8. \u5317\u6d77\u6c49\uff08\u5317\u6d77\u6c49\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77\u6c49\uff08\u5317\u6d77\u6c49\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n9. \u5317\u6d77aut\uff08\u5317\u6d77aut\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77aut\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n10. \u5317\u6d77aut\uff08\u5317\u6d77aut\uff09\uff1a\u5317\u6d77aut\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77aut\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n11. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77aut\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n12. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u4e09\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n13. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n14. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n15. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n16. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n17. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n18. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n19. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n20. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n21. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n22. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n23. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n24. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n25. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n26. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n27. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n28. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n29. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n30. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n31. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77aut\uff08\u5317\u6d77\u9053\uff09\u306b\u9762\u3057\u3066\u3044\u307e\u3059\u3002\n\n32. \u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\uff1a\u5317\u6d77\u9053\uff08\u5317\u6d77\u9053\uff09\u306e\u4e2d\u592e\u90e8\u5206\u306f\u3001\u5317\u6d77\n<\/code><\/pre>\n<p><em>(Cut off at the 1024-token limit.)<\/em><\/p>\n<p>Reference: The correct answer is eight prefectures: Tochigi, Gunma, Saitama, Yamanashi, Nagano, Gifu, Shiga and Nara.<\/p>\n<p><strong>Constrained writing<\/strong><\/p>\n<blockquote>\n<p>\u300c\u6625\u300d\u3092\u30c6\u30fc\u30de\u306b\u3057\u305f\u4ff3\u53e5\u30923\u3064\u3001\u4e94\u30fb\u4e03\u30fb\u4e94\u306e\u97f3\u6570\u3092\u5b88\u3063\u3066\u4f5c\u3063\u3066\u304f\u3060\u3055\u3044\u3002\u305d\u308c\u305e\u308c\u306b\u4e00\u8a00\u306e\u8aac\u660e\u3092\u4ed8\u3051\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n<\/blockquote>\n<p><em>(Write three haiku on the theme of spring, keeping the 5-7-5 syllable pattern, each with a one-line explanation.)<\/em><\/p>\n<p><code>F16<\/code> (baseline) 8.22GB<\/p>\n<pre><code class=\"language-text\">1. \u96e8\u4e0a\u304c\u308a\u306e\u7a7a  \n\u6625\u306e\u606f\u5439\u304c\u5c4a\u304f\n\n2. \u82b1\u898b\u306e\u91cc  \n\u6625\u306f\u5fae\u7b11\u307f\u8fd4\u3059\n\n3. \u96ea\u89e3\u3051\u306e\u5ddd  \n\u6625\u306e\u97f3\u304c\u805e\u3053\u3048\u308b\n\n\u3010\u8aac\u660e\u3011  \n1. \u96e8\u4e0a\u304c\u308a\u306e\u7a7a \u2014 \u6625\u306e\u59cb\u307e\u308a\u3092\u8c61\u5fb4\u3059\u308b\u7a7a\u306e\u5909\u5316\u3002  \n2. \u82b1\u898b\u306e\u91cc \u2014 \u6625\u306e\u4ee3\u8868\u7684\u306a\u884c\u4e8b\u300c\u82b1\u898b\u300d\u3092\u63cf\u5199\u3002  \n3. \u96ea\u89e3\u3051\u306e\u5ddd \u2014 \u51ac\u306e\u7d42\u308f\u308a\u3068\u6625\u306e\u59cb\u307e\u308a\u306e\u79fb\u884c\u3092\u8868\u3059\u3002\n<\/code><\/pre>\n<p><code>Q4_K_M<\/code> 2.53GB<\/p>\n<pre><code class=\"language-text\">1. \u96e8\u4e0a\u304c\u308a\u306e\u7a7a  \n\u6625\u306e\u5302\u3044\u304c  \n\u67d3\u307f\u3066\u6765\u30eb\n\n\u3010\u89e3\u8aac\u3011\u96e8\u4e0a\u304c\u308a\u306e\u7a7a\u306b\u6625\u306e\u5302\u3044\u304c\u67d3\u307f\u3066\u6765\u308b\u3053\u3068\u3067\u3001\u6625\u306e\u8a2a\u308c\u3068\u7a7a\u6c17\u306e\u5909\u5316\u3092\u63cf\u5199\u3002\n\n2. \u82b1\u898b\u306e\u65e5  \n\u8776\u3005\u821e\u3044\u6563\u308b  \n\u5fc3\u306b\u6b8b\u308b\n\n\u3010\u89e3\u8aac\u3011\u82b1\u898b\u306e\u65e5\u3001\u8776\u3005\u304c\u821e\u3044\u6563\u308b\u69d8\u5b50\u3092\u63cf\u304d\u3001\u6625\u306e\u7f8e\u3057\u3055\u3068\u5fc3\u306b\u6b8b\u308b\u611f\u52d5\u3092\u8868\u73fe\u3002\n\n3. \u671d\u65e5\u304c\u6607\u308b  \n\u8349\u306e\u82bd\u304c\u81a8\u3089  \n\u6625\u306e\u59cb\u307e\u308a\n\n\u3010\u89e3\u8aac\u3011\u671d\u65e5\u304c\u6607\u308a\u3001\u8349\u306e\u82bd\u304c\u81a8\u3089\u3080\u6625\u306e\u59cb\u307e\u308a\u3092\u8c61\u5fb4\u7684\u306b\u63cf\u5199\u3002\n<\/code><\/pre>\n<p><code>Q2_K<\/code> 1.67GB<\/p>\n<pre><code class=\"language-text\">\u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n1. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n2. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n3. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n4. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n5. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n6. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n7. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n8. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n9. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n10. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n11. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n12. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n13. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n14. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n15. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n16. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n17. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n18. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n19. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n20. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n21. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n22. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n23. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n24. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n25. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n26. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n27. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n28. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n29. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n30. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n31. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n32. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n33. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n34. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n35. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n36. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n37. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n38. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n39. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n40. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n41. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449\u3002\n\n42. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n43. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n44. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n45. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n46. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n50. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n51. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n52. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n53. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n54. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n55. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n56. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n57. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n58. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n59. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n60. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n61. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n62. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n63. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n64. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n65. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n66. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n67. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n68. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n69. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n\n70. \u6625\u306f\u3001\u81ea\u7136\u306e\u8a00\u8449.\n71. \u7d76\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n72. \u7d76\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n73. \u7d76\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n74. \u7d5e\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n75. \u7d6e\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n76. \u7d71\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n77. \u7d71\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n78. \u7d71\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n79. \u7d71\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n80. \u7d71\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n81. \u7d71\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n82. \u7d71\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n83. \u7d71\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n84. \u7d71\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n85. \u7d71\u547d\u306e\u89e3\u6cd5(\u89e3\u6cd5)\n\n86. \u7d71\u547d\u306e\u89e3\u6cd5(\n<\/code><\/pre>\n<p><em>(Cut off at the 1024-token limit.)<\/em><\/p>\n<h3>Running It on a CPU Only<\/h3>\n<div class=\"lmw-table-scroll\" tabindex=\"0\" style=\"overflow-x:auto;-webkit-overflow-scrolling:touch;max-width:100%;\">\n<table style=\"width:max-content;min-width:100%;border-collapse:collapse;\">\n<thead>\n<tr>\n<th>Item<\/th>\n<th>Result<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Prompt processing (512 tokens)<\/td>\n<td>19.4 tokens\/s<\/td>\n<\/tr>\n<tr>\n<td>Generation (128 tokens)<\/td>\n<td>8.3 tokens\/s<\/td>\n<\/tr>\n<tr>\n<td>Generation in Japanese characters<\/td>\n<td>about 12.1 characters\/s<\/td>\n<\/tr>\n<tr>\n<td>Peak memory (context 4,096 tokens)<\/td>\n<td>6.7GB<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p>Conditions: Neoverse-N1, 3 threads, no GPU, llama.cpp b11223, <code>Q4_K_M<\/code> (2.53GB) from <a href=\"https:\/\/huggingface.co\/mradermacher\/WeVisDoc-4B-GGUF\">mradermacher\/WeVisDoc-4B-GGUF<\/a>, measured 2026-09-29 (JST). Other models measured under the same conditions (closest in file size):<\/p>\n<div class=\"lmw-table-scroll\" tabindex=\"0\" style=\"overflow-x:auto;-webkit-overflow-scrolling:touch;max-width:100%;\">\n<p class=\"lmw-table-hint\" style=\"margin:0 0 4px;font-size:0.85em;opacity:0.7;\">\u2192 Scroll horizontally to see all columns<\/p>\n<table style=\"width:max-content;min-width:100%;border-collapse:collapse;\">\n<thead>\n<tr>\n<th>Model<\/th>\n<th>Quant<\/th>\n<th>File size<\/th>\n<th>Generation<\/th>\n<th>In Japanese<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Qwen3-4B<\/td>\n<td><code>Q4_K_M<\/code><\/td>\n<td>2.33GB<\/td>\n<td>6.8 tokens\/s<\/td>\n<td>\u2014<\/td>\n<\/tr>\n<tr>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/08\/minicpm5-2b-released\/\">MiniCPM5-2B<\/a><\/td>\n<td><code>Q4_K_M<\/code><\/td>\n<td>1.45GB<\/td>\n<td>13.9 tokens\/s<\/td>\n<td>20.9 chars\/s<\/td>\n<\/tr>\n<tr>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/16\/eleutherai-olmo-3-7b-reward-hacking-models\/\">olmo3-7b-sdf-sft-clean150<\/a><\/td>\n<td><code>Q4_K_M<\/code><\/td>\n<td>4.16GB<\/td>\n<td>4.9 tokens\/s<\/td>\n<td>4.4 chars\/s<\/td>\n<\/tr>\n<tr>\n<td>Qwen3-8B<\/td>\n<td><code>Q4_K_M<\/code><\/td>\n<td>4.68GB<\/td>\n<td>4.6 tokens\/s<\/td>\n<td>\u2014<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p>Peak memory includes the model file mapped into memory (mmap). A PC with a GPU will be much faster; these figures are a guide for CPU-only machines such as mini PCs and ARM boards.<\/p>\n<h3>Answers About Images (Verbatim)<\/h3>\n<p>Under the same conditions, we gave the model images we drew ourselves with code (a chart and a table in Japanese) and asked about them. Answers are unedited; the correct answers were computed by the code that drew the images.<\/p>\n<p><strong>Reading a chart<\/strong><\/p>\n<p><img decoding=\"async\" alt=\"Reading a chart\" src=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/wp-content\/uploads\/2026\/09\/lmw-lab-vision-chart-v1.png\" \/><\/p>\n<blockquote>\n<p>\u3053\u306e\u30b0\u30e9\u30d5\u3067\u6765\u5ba2\u6570\u304c\u6700\u3082\u591a\u3044\u66dc\u65e5\u3068\u3001\u305d\u306e\u4eba\u6570\u3092\u7b54\u3048\u3066\u304f\u3060\u3055\u3044\u3002\u307e\u305f\u3001\u5e73\u65e5(\u6708\u66dc\u65e5\u301c\u91d1\u66dc\u65e5)\u306e\u6765\u5ba2\u6570\u306e\u5408\u8a08\u3082\u6c42\u3081\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n<\/blockquote>\n<p><em>(Which day has the most visitors in this chart, and how many? Also give the weekday (Mon-Fri) total.)<\/em><\/p>\n<pre><code class=\"language-text\">\u3042\u308b\u66f8\u5e97\u306e\u66dc\u65e5\u5225\u306e\u6765\u5ba2\u6570(1\u9031\u9593)\n\n\u6700\u3082\u591a\u3044\u66dc\u65e5: \u571f\u66dc\u65e5\n\u4eba\u6570: 260\u4eba\n\n\u5e73\u65e5\u306e\u6765\u5ba2\u6570\u306e\u5408\u8a08: 695\u4eba\n<\/code><\/pre>\n<p>Reference: Correct answer: Sat with 260 visitors; the weekday total is 635.<\/p>\n<p><strong>Reading a table and calculating<\/strong><\/p>\n<p><img decoding=\"async\" alt=\"Reading a table and calculating\" src=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/wp-content\/uploads\/2026\/09\/lmw-lab-vision-receipt-v1.png\" \/><\/p>\n<blockquote>\n<p>\u3053\u306e\u8868\u306e\u54c1\u76ee\u3054\u3068\u306e\u91d1\u984d(\u6570\u91cf\u00d7\u5358\u4fa1)\u3068\u3001\u5408\u8a08\u91d1\u984d\u3092\u7b54\u3048\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n<\/blockquote>\n<p><em>(Give the amount for each item (quantity \u00d7 unit price) in this table, and the total.)<\/em><\/p>\n<pre><code class=\"language-text\">## \u8cb7\u3044\u7269\u306e\u660e\u7d30\n\n&lt;table&gt;&lt;tr&gt;&lt;td&gt;\u54c1\u76ee&lt;\/td&gt;&lt;td&gt;\u6570\u91cf&lt;\/td&gt;&lt;td&gt;\u5358\u4fa1&lt;\/td&gt;&lt;\/tr&gt;&lt;tr&gt;&lt;td&gt;\u308a\u3093\u3054&lt;\/td&gt;&lt;td&gt;3&lt;\/td&gt;&lt;td&gt;120\u5186&lt;\/td&gt;&lt;\/tr&gt;&lt;tr&gt;&lt;td&gt;\u307f\u304b\u3093&lt;\/td&gt;&lt;td&gt;8&lt;\/td&gt;&lt;td&gt;45\u5186&lt;\/td&gt;&lt;\/tr&gt;&lt;tr&gt;&lt;td&gt;\u725b\u4e73&lt;\/td&gt;&lt;td&gt;2&lt;\/td&gt;&lt;td&gt;198\u5186&lt;\/td&gt;&lt;\/tr&gt;&lt;tr&gt;&lt;td&gt;\u98df\u30d1\u30f3&lt;\/td&gt;&lt;td&gt;1&lt;\/td&gt;&lt;td&gt;238\u5186&lt;\/td&gt;&lt;\/tr&gt;&lt;\/table&gt;\n\n&lt;table&gt;&lt;tr&gt;&lt;td&gt;\u54c1\u76ee&lt;\/td&gt;&lt;td&gt;\u6570\u91cf&lt;\/td&gt;&lt;td&gt;\u5358\u4fa1&lt;\/td&gt;&lt;\/tr&gt;&lt;tr&gt;&lt;td&gt;\u308a\u3093\u3054&lt;\/td&gt;&lt;td&gt;3&lt;\/td&gt;&lt;td&gt;120\u5186&lt;\/td&gt;&lt;\/tr&gt;&lt;tr&gt;&lt;td&gt;\u307f\u304b\u3093&lt;\/td&gt;&lt;td&gt;8&lt;\/td&gt;&lt;td&gt;45\u5186&lt;\/td&gt;&lt;\/tr&gt;&lt;tr&gt;&lt;td&gt;\u725b\u4e73&lt;\/td&gt;&lt;td&gt;2&lt;\/td&gt;&lt;td&gt;198\u5186&lt;\/td&gt;&lt;\/tr&gt;&lt;tr&gt;&lt;td&gt;\u98df\u30d1\u30f3&lt;\/td&gt;&lt;td&gt;1&lt;\/td&gt;&lt;td&gt;238\u5186&lt;\/td&gt;&lt;\/tr&gt;&lt;\/table&gt;\n\n\u5408\u8a08\u91d1\u984d: 1,009\u5186\n<\/code><\/pre>\n<p>Reference: Correct answer: 360 yen, 360 yen, 396 yen, 238 yen; total 1,354 yen.<\/p>\n<p><!-- \/lmw:lab --><\/p>\n<h2>How to Get It<\/h2>\n<p>Model files are distributed in safetensors format and can be obtained directly from Hugging Face.<\/p>\n<p>An example script execution for starting a server using vLLM (&gt;=0.11.1) is as follows:<\/p>\n<pre><code class=\"language-bash\">bash scripts\/serve_vllm.sh Tencent\/WeVisDoc-2B\n<\/code><\/pre>\n<p>*When using the 4B model, specify <code>Tencent\/WeVisDoc-4B<\/code>.<\/p>\n<p>After starting the service, here is an example command to parse an image using the included client:<\/p>\n<pre><code class=\"language-bash\">python -m wevisdoc.client --image page.png --output results\/page.md\n<\/code><\/pre>\n<p>It can also be executed directly in a local inference environment using Transformers.<\/p>\n<pre><code class=\"language-bash\">python -m wevisdoc.local --model Tencent\/WeVisDoc-2B --image page.png --output results\/page.md\n<\/code><\/pre>\n<p><!-- lmw:variants --><\/p>\n<h2>Quantized and Converted Variants<\/h2>\n<div class=\"lmw-table-scroll\" tabindex=\"0\" style=\"overflow-x:auto;-webkit-overflow-scrolling:touch;max-width:100%;\">\n<p class=\"lmw-table-hint\" style=\"margin:0 0 4px;font-size:0.85em;opacity:0.7;\">\u2192 Scroll horizontally to see all columns<\/p>\n<table style=\"width:max-content;min-width:100%;border-collapse:collapse;\">\n<thead>\n<tr>\n<th>Added<\/th>\n<th>Publisher<\/th>\n<th>Format<\/th>\n<th>Repository<\/th>\n<th>Smallest VRAM tier (build, est. memory)<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>2026-09-20 (2 days after publication)<\/td>\n<td>mradermacher<\/td>\n<td>GGUF<\/td>\n<td><a href=\"https:\/\/huggingface.co\/mradermacher\/WeVisDoc-4B-GGUF\">mradermacher\/WeVisDoc-4B-GGUF<\/a><\/td>\n<td>Q5_K_S 3.5GB (fits in 4GB VRAM)<\/td>\n<\/tr>\n<tr>\n<td>2026-09-20 (2 days after publication)<\/td>\n<td>mradermacher<\/td>\n<td>GGUF (imatrix)<\/td>\n<td><a href=\"https:\/\/huggingface.co\/mradermacher\/WeVisDoc-4B-i1-GGUF\">mradermacher\/WeVisDoc-4B-i1-GGUF<\/a><\/td>\n<td>Q5_K_S 3.5GB (fits in 4GB VRAM)<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p>File sizes of each build:<\/p>\n<ul>\n<li>Available builds in mradermacher\/WeVisDoc-4B-GGUF: Q2_K 1.7GB \/ Q3_K_S 1.9GB \/ Q3_K_M 2.1GB \/ Q3_K_L 2.2GB \/ IQ4_XS 2.3GB \/ Q4_K_S 2.4GB \/ Q4_K_M 2.5GB \/ Q5_K_S 2.9GB \/ Q5_K_M 2.9GB \/ Q6_K 3.4GB \/ Q8_0 4.4GB \/ F16 8.2GB<\/li>\n<li>Available builds in mradermacher\/WeVisDoc-4B-i1-GGUF: IQ1_S 1.1GB \/ IQ1_M 1.2GB \/ IQ2_XXS 1.3GB \/ IQ2_XS 1.4GB \/ IQ2_S 1.5GB \/ IQ2_M 1.6GB \/ Q2_K_S 1.6GB \/ Q2_K 1.7GB \/ IQ3_XXS 1.7GB \/ IQ3_XS 1.8GB \/ Q3_K_S 1.9GB \/ IQ3_S 1.9GB \/ IQ3_M 2.0GB \/ Q3_K_M 2.1GB \/ Q3_K_L 2.2GB \/ IQ4_XS 2.3GB \/ Q4_0 2.4GB \/ IQ4_NL 2.4GB \/ Q4_K_S 2.4GB \/ Q4_K_M 2.5GB \/ Q4_1 2.6GB \/ Q5_K_S 2.9GB \/ Q5_K_M 2.9GB \/ Q6_K 3.4GB<\/li>\n<\/ul>\n<p>In addition, 1 converted build(s) from other uploaders exist on Hugging Face; this site lists only builds from the model&#8217;s publisher or established quantization maintainers.<\/p>\n<p><em>This section is appended automatically by Local Model Watch when a converted build of this model appears after publication. Memory figures are estimated from the size of the distributed files. What the quantization names mean: <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/glossary-quantization-en\/\">glossary<\/a>.<\/em><\/p>\n<p><!-- \/lmw:variants --><\/p>\n<p><!-- lmw:related --><\/p>\n<h2>Related Articles<\/h2>\n<ul>\n<li><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/14\/tencent-simple-attention-sparsification-qwen3\/\">Simple-Attention-Sparsification Text Generation Model: 12GB+ VRAM<\/a><\/li>\n<\/ul>\n<p><!-- \/lmw:related --><\/p>\n<p><!-- lmw:next-steps --><\/p>\n<h2>What to Read Next<\/h2>\n<ul>\n<li><strong>Find models by VRAM<\/strong> (This model runs from the 4GB tier) \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/vram-guide-8gb-en\/\">Other models that run on a 8GB GPU<\/a><\/li>\n<li><strong>Explore the same model family<\/strong> \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/model-tencent-wevisdoc-en\/\">WeVisDoc family overview (1 articles, 2 converted builds)<\/a><\/li>\n<li><strong>What Q5_K_S, Q8_0 mean and where to get this model<\/strong> \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/format-gguf-en\/\">GGUF format guide and models<\/a> \/ <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/glossary-quantization-en\/\">Quantization and model-format glossary<\/a><\/li>\n<li><strong>Learn about the publisher<\/strong> \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/publisher-tencent-en\/\">Tencent: models, licenses and articles<\/a><\/li>\n<\/ul>\n<p><!-- \/lmw:next-steps --><\/p>\n<h2>Sources<\/h2>\n<ul>\n<li><a href=\"https:\/\/huggingface.co\/tencent\/WeVisDoc-4B\">https:\/\/huggingface.co\/tencent\/WeVisDoc-4B<\/a><\/li>\n<li><a href=\"https:\/\/huggingface.co\/tencent\/WeVisDoc-2B\">https:\/\/huggingface.co\/tencent\/WeVisDoc-2B<\/a><\/li>\n<\/ul>\n<p><!-- lmw:updates --><\/p>\n<h2>Update History<\/h2>\n<ul>\n<li>2026-09-20: Added converted builds to \u201cQuantized and Converted Variants\u201d: mradermacher\/WeVisDoc-4B-GGUF, mradermacher\/WeVisDoc-4B-i1-GGUF<\/li>\n<li>2026-09-20: Updated the hardware requirements table with the actual file sizes of mradermacher\/WeVisDoc-4B-GGUF.<\/li>\n<li>2026-09-24: Rebuilt the article: Added a &#8220;can you run it locally?&#8221; verdict from the distribution formats, converted builds, engine registries, license terms and measured compression<\/li>\n<li>2026-09-29: Added our own measurements: Japanese token efficiency, what is inside the GGUF file.<\/li>\n<li>2026-09-29: Added our own measurements: CPU speed and answers to Japanese questions.<\/li>\n<li>2026-09-29: Added our own measurements: quality loss by quantization level.<\/li>\n<\/ul>\n<p><!-- \/lmw:updates --><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Tencent has released WeVisDoc-2B and WeVisDoc-4B, end-to-end document parsing models based on Qwen3-VL that extract structured text from page images.<\/p>\n","protected":false},"author":1,"featured_media":1506,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[310],"tags":[1490,1492,1494,1172,1496,117],"class_list":["post-1507","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-new-models","tag-document-parsing-en","tag-ocr-en","tag-qwen3-vl-en","tag-tencent-en","tag-wevisdoc-en","tag--en"],"lang":"en","translations":{"en":1507,"ja":1505},"_links":{"self":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/posts\/1507","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/comments?post=1507"}],"version-history":[{"count":60,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/posts\/1507\/revisions"}],"predecessor-version":[{"id":14829,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/posts\/1507\/revisions\/14829"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/media\/1506"}],"wp:attachment":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/media?parent=1507"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/categories?post=1507"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/tags?post=1507"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}