{"id":354,"date":"2026-09-09T15:19:05","date_gmt":"2026-09-09T06:19:05","guid":{"rendered":"https:\/\/localmodelwatch.tsuchitsuchi.com\/2026\/09\/09\/nex-n25-mini-released\/"},"modified":"2026-11-03T16:08:18","modified_gmt":"2026-11-03T07:08:18","slug":"nex-n25-mini-released","status":"publish","type":"post","link":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/09\/nex-n25-mini-released\/","title":{"rendered":"Nex-N2.5-mini Text Generation Model: Our Test Answers, 16GB+ VRAM"},"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\/nex-agi\/Nex-N2.5-mini\">nex-agi\/Nex-N2.5-mini<\/a><\/td>\n<\/tr>\n<tr>\n<td>Family guide<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/model-nex-agi-nex-n2-5-en\/\">Nex-N2.5 guide (2 articles)<\/a><\/td>\n<\/tr>\n<tr>\n<td>Publisher guide<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/publisher-nex-agi-en\/\">Nex-AGI: models and licenses<\/a><\/td>\n<\/tr>\n<tr>\n<td>Published<\/td>\n<td>2026-09-08<\/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>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>nex-agi_Nex-N2.5-mini-Q2_K.gguf<\/code> (Q2_K) on our CPU-only server (Neoverse-N1, 3 threads, llama.cpp b11223): generation at 7.3 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>It needs about 21% fewer tokens than the Qwen3 tokenizer for the same Japanese text, so about 1.26\u00d7 as much Japanese fits in the same context length, and generation is faster per character.<\/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>Nex-AGI has released &#8220;Nex-N2.5-mini&#8221;, a multimodal foundational model from its next-generation agent model family &#8220;Nex-N2.5&#8221; designed for long-running tasks in real-world environments. It focuses on computer operation, web browsing, and self-correction through visual feedback, and is available as an open-weight model.<\/p>\n<h2>Specifications<\/h2>\n<ul>\n<li>License: <code>apache-2.0<\/code><\/li>\n<li>Architecture: <code>Qwen3_5MoeForConditionalGeneration<\/code> (<code>qwen3_5_moe<\/code>)<\/li>\n<\/ul>\n<h2>Performance<\/h2>\n<p>Comparison tables for text and multimodal benchmarks provided in the model card are shown below. The comparisons include &#8220;Nex-N2.5-Pro&#8221;, &#8220;Nex-N2.5-Max&#8221;, and other major models.<\/p>\n<h3>Text Benchmarks<\/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>Benchmark \/ CODING 3<\/th>\n<th>Nex-N2.5-mini \/ CODING 3<\/th>\n<th>Nex-N2.5-Pro \/ CODING 3<\/th>\n<th>Nex-N2.5-Max \/ CODING 3<\/th>\n<th>Claude Opus 5 \/ CODING 3<\/th>\n<th>GPT-5.6 Sol \/ CODING 3<\/th>\n<th>Kimi-K3 \/ CODING 3<\/th>\n<th>GLM-5.3 \/ CODING 3<\/th>\n<th>DeepSeek-V4-Pro-0813 4 \/ CODING 3<\/th>\n<th>Qwen3.8-Max \/ CODING 3<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>Terminal-Bench 2.1<\/td>\n<td>73.4<\/td>\n<td>82.7<\/td>\n<td>86.1<\/td>\n<td>89.1<\/td>\n<td>88.8<\/td>\n<td>88.3<\/td>\n<td>88.2<\/td>\n<td>87.9<\/td>\n<td>86.6<\/td>\n<\/tr>\n<tr>\n<td>SWE-Bench Pro<\/td>\n<td>43.8<\/td>\n<td>61.2<\/td>\n<td>65.7<\/td>\n<td>79.2<\/td>\n<td>64.6<\/td>\n<td>63.3<\/td>\n<td>64.6<\/td>\n<td>55.4<\/td>\n<td>67.7<\/td>\n<\/tr>\n<tr>\n<td>DeepSWE v1.1<\/td>\n<td>36.1<\/td>\n<td>55.8<\/td>\n<td>65.6<\/td>\n<td>73.7<\/td>\n<td>72.7<\/td>\n<td>67.5<\/td>\n<td>66.9<\/td>\n<td>62.8<\/td>\n<td>69.3<\/td>\n<\/tr>\n<tr>\n<td>AGENTIC<\/td>\n<td>AGENTIC<\/td>\n<td>AGENTIC<\/td>\n<td>AGENTIC<\/td>\n<td>AGENTIC<\/td>\n<td>AGENTIC<\/td>\n<td>AGENTIC<\/td>\n<td>AGENTIC<\/td>\n<td>AGENTIC<\/td>\n<td>AGENTIC<\/td>\n<\/tr>\n<tr>\n<td>AutomationBench v1.0.6 5<\/td>\n<td>32.3<\/td>\n<td>44.2<\/td>\n<td>50.2<\/td>\n<td>50.3<\/td>\n<td>45.8<\/td>\n<td>46.7<\/td>\n<td>48.2<\/td>\n<td>43.2<\/td>\n<td>39.8<\/td>\n<\/tr>\n<tr>\n<td>Toolathlon Verified<\/td>\n<td>54.6<\/td>\n<td>68.5<\/td>\n<td>74.7<\/td>\n<td>76.5<\/td>\n<td>74.9<\/td>\n<td>76.5<\/td>\n<td>73.0<\/td>\n<td>74.1<\/td>\n<td>72.5<\/td>\n<\/tr>\n<tr>\n<td>GDPval-AA v2<\/td>\n<td>1446<\/td>\n<td>1628<\/td>\n<td>1713<\/td>\n<td>1831<\/td>\n<td>1711<\/td>\n<td>1675<\/td>\n<td>1763<\/td>\n<td>1580<\/td>\n<td>1717<\/td>\n<\/tr>\n<tr>\n<td>Job Bench<\/td>\n<td>28.5<\/td>\n<td>41.4<\/td>\n<td>53.6<\/td>\n<td>65.7<\/td>\n<td>45.4<\/td>\n<td>52.9<\/td>\n<td>58.2<\/td>\n<td>54.1<\/td>\n<td>53.4<\/td>\n<\/tr>\n<tr>\n<td>BrowseComp 6<\/td>\n<td>83.4<\/td>\n<td>89.7<\/td>\n<td>92.6<\/td>\n<td>90.8<\/td>\n<td>90.4<\/td>\n<td>91.2<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<h3>Multimodal Benchmarks<\/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>Benchmark<\/th>\n<th>Nex-N2.5-mini<\/th>\n<th>Nex-N2.5-Pro<\/th>\n<th>MiniMax-M3<\/th>\n<th>Claude Opus 5<\/th>\n<th>GPT-5.6 Sol<\/th>\n<th>Kimi-K3<\/th>\n<th>GLM-5.3-Flash<\/th>\n<th>DeepSeek-V4-Flash-Vision<\/th>\n<th>Qwen3.8-Max<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>OSWorld-Verified 8<\/td>\n<td>71.2<\/td>\n<td>82.2<\/td>\n<td>75.2<\/td>\n<td>83.4<\/td>\n<td>83.2<\/td>\n<td>84.8<\/td>\n<td>62.3<\/td>\n<td>76.7<\/td>\n<td>86.1<\/td>\n<\/tr>\n<tr>\n<td>OSWorld-2<\/td>\n<td>30.5<\/td>\n<td>56.4<\/td>\n<td>22.3<\/td>\n<td>68.3<\/td>\n<td>62.7<\/td>\n<td>58.3<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>46.7<\/td>\n<\/tr>\n<tr>\n<td>WebTest 8, 9<\/td>\n<td>48.6<\/td>\n<td>52.8<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>54.0<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>52.3<\/td>\n<\/tr>\n<tr>\n<td>WebArena-Verified 8<\/td>\n<td>63.4<\/td>\n<td>67.6<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>69.7<\/td>\n<td>71.6<\/td>\n<td>\u2014<\/td>\n<td>62.3<\/td>\n<td>66.8<\/td>\n<\/tr>\n<tr>\n<td>OSWorld-G<\/td>\n<td>82.9<\/td>\n<td>87.4<\/td>\n<td>\u2014<\/td>\n<td>76.8<\/td>\n<td>77.7<\/td>\n<td>79.6<\/td>\n<td>83.3<\/td>\n<td>59.4<\/td>\n<td>84.9<\/td>\n<\/tr>\n<tr>\n<td>Vision2Web 7<\/td>\n<td>52.9<\/td>\n<td>68.2<\/td>\n<td>59.0<\/td>\n<td>\u2014<\/td>\n<td>79.8<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>75.1<\/td>\n<\/tr>\n<tr>\n<td>SWE-MM<\/td>\n<td>25.5<\/td>\n<td>38.2<\/td>\n<td>\u2014<\/td>\n<td>59.4<\/td>\n<td>40.2<\/td>\n<td>37.3<\/td>\n<td>20.6<\/td>\n<td>39.2<\/td>\n<td>39.2<\/td>\n<\/tr>\n<tr>\n<td>OmniDoc<\/td>\n<td>89.7<\/td>\n<td>92.2<\/td>\n<td>91.6<\/td>\n<td>\u2014<\/td>\n<td>92.9<\/td>\n<td>91.1<\/td>\n<td>\u2014<\/td>\n<td>\u2014<\/td>\n<td>92.1<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p>According to the evaluation results from the publishers, Nex-N2.5-mini is the lightest model in the family, and its scores tend to be generally lower compared to the higher-tier Pro and Max models as well as large models from other companies. On the other hand, it demonstrates solid performance in certain tasks, scoring 73.4 on Terminal-Bench 2.1 and 83.4 on BrowseComp. In multimodal benchmarks such as OSWorld-Verified and OmniDoc, while its numbers are lower than the Pro model in the family and other models, it maintains a certain standard.<\/p>\n<h2>Strengths and Use Cases<\/h2>\n<p>It is built with a focus on operating as an agent for computer operations, web browsing, and visual grounding. A key characteristic is that it utilizes vision not merely as an input modality, but as a critical interface for perceiving the environment, verifying results, and progressing through tasks.<\/p>\n<p><!-- lmw:hardware --><\/p>\n<h2>Hardware Requirements<\/h2>\n<p><strong>Estimated requirements (calculated by Local Model Watch)<\/strong> \u2014 35.1B 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>16GB (RTX 5060 Ti 16GB \/ 4060 Ti 16GB, etc.)<\/td>\n<td>Q2_K<\/td>\n<td>12.9GB<\/td>\n<td>15.5GB<\/td>\n<\/tr>\n<tr>\n<td>24GB (RTX 4090 \/ 3090, etc.)<\/td>\n<td>Q4_K_S<\/td>\n<td>19.5GB<\/td>\n<td>23.4GB<\/td>\n<\/tr>\n<tr>\n<td>32GB (RTX 5090, etc.)<\/td>\n<td>Q5_K_M<\/td>\n<td>25.1GB<\/td>\n<td>30.2GB<\/td>\n<\/tr>\n<tr>\n<td>48GB (RTX 6000 Ada \/ A6000, etc.)<\/td>\n<td>Q8_0<\/td>\n<td>34.4GB<\/td>\n<td>41.3GB<\/td>\n<\/tr>\n<tr>\n<td>80GB class (A100 \/ H100)<\/td>\n<td>BF16<\/td>\n<td>64.6GB<\/td>\n<td>77.5GB<\/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\/bartowski\/nex-agi_Nex-N2.5-mini-GGUF\">bartowski\/nex-agi_Nex-N2.5-mini-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><strong>Runs in Ollama, LM Studio and llama.cpp via a converted build.<\/strong><\/p>\n<p>The publisher ships safetensors, but <a href=\"https:\/\/huggingface.co\/bartowski\/nex-agi_Nex-N2.5-mini-GGUF\">bartowski\/nex-agi_Nex-N2.5-mini-GGUF<\/a> provides a GGUF build you can use.<\/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 Q2_K build measures 3.16 bits per weight \u2014 about 20% 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, converted builds we have found, and each engine&#8217;s own model registry. &#8220;Not found&#8221; means we have not seen such a build, not that none exists. 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 nex-agi&#8217;s Nex-N2.5-mini solely on a CPU, we found that it achieved good Japanese token efficiency and smoothly answered tasks such as word problems and summarization, while failing to generate haiku and hitting the upper token limit with truncation. During verification in a CPU environment using the Q2_K quantized version, the generation speed reached 7.31 tokens per second, translating to 13.4 characters per second in Japanese output. This speed is fast compared to the other models measured under the same conditions, though users need to note that peak memory usage reached 14.9 GB.<\/p>\n<p>Regarding Japanese text processing, the number of tokens per 1,000 characters is 546, which allows more Japanese text to fit into the context with fewer tokens compared to Qwen3 at 688 tokens. In the five-question test, the model successfully generated responses for drafting a polite business email, solving a shopping word problem with the correct answer of 280 yen, and accurately summarizing a 1,200-character article into exactly three sentences. For the knowledge question, it correctly listed seven out of the eight prefectures that do not face the sea.<\/p>\n<p>On the other hand, the instruction to create three spring haiku resulted in no answer, hitting the 1,024-token upper limit and cutting off mid-generation. Additionally, the knowledge question included one prefecture that actually faces the sea by mistake. Overall, while the model demonstrates strong token efficiency and practical text generation capabilities, it also exhibits weaknesses such as truncation in specific creative tasks and occasional inaccuracies in factual knowledge.<\/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>546<\/strong><\/td>\n<td><strong>0.99\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>It needs about 21% fewer tokens than the Qwen3 tokenizer for the same Japanese text, so about 1.26\u00d7 as much Japanese fits in the same context length, and generation is faster per character.<\/p>\n<p>Counted with the <code>tokenizer.json<\/code> of <a href=\"https:\/\/huggingface.co\/nex-agi\/Nex-N2.5-mini\">nex-agi\/Nex-N2.5-mini<\/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\/bartowski\/nex-agi_Nex-N2.5-mini-GGUF\/blob\/main\/nex-agi_Nex-N2.5-mini-Q4_K_M.gguf\">nex-agi_Nex-N2.5-mini-Q4_K_M.gguf<\/a> (20.79GB, <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>qwen35moe<\/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>40<\/td>\n<\/tr>\n<tr>\n<td>Experts<\/td>\n<td>8 active out of 256<\/td>\n<\/tr>\n<tr>\n<td>Vocabulary size<\/td>\n<td>248,320<\/td>\n<\/tr>\n<tr>\n<td>Chat template<\/td>\n<td>Included (mentions tool calls, has a thinking switch)<\/td>\n<\/tr>\n<tr>\n<td>imatrix<\/td>\n<td>Used (<code>nex-agi_Nex-N2.5-mini-calibration-v6.txt<\/code>, 573 chunks)<\/td>\n<\/tr>\n<tr>\n<td>Weight types (share of parameters)<\/td>\n<td>Q4_K 75.9% \/ Q6_K 12.4% \/ Q8_0 8.5% \/ Q5_K 3.1% \/ other 0.1%<\/td>\n<\/tr>\n<tr>\n<td>Average bits per weight<\/td>\n<td>5.15 bits<\/td>\n<\/tr>\n<tr>\n<td>Embedding \/ output layer type<\/td>\n<td>Q8_0 \/ Q8_0<\/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>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>15.7 tokens\/s<\/td>\n<\/tr>\n<tr>\n<td>Generation (128 tokens)<\/td>\n<td>7.3 tokens\/s<\/td>\n<\/tr>\n<tr>\n<td>Generation in Japanese characters<\/td>\n<td>about 13.4 characters\/s<\/td>\n<\/tr>\n<tr>\n<td>Peak memory (context 4,096 tokens)<\/td>\n<td>14.9GB<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p>Conditions: Neoverse-N1, 3 threads, no GPU, llama.cpp b11223, <code>Q2_K<\/code> (12.89GB) from <a href=\"https:\/\/huggingface.co\/bartowski\/nex-agi_Nex-N2.5-mini-GGUF\">bartowski\/nex-agi_Nex-N2.5-mini-GGUF<\/a>, measured 2026-09-29 (JST). The file plus the part that reads images would exceed our 13GB limit, so we skipped the image questions and asked only the text questions. The 4-bit files did not fit our 13GB limit, so we measured the lower-bit <code>Q2_K<\/code> file; quality is lower than at 4 bits. 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>K2-Horizon-MoVA-36B-A4B<\/td>\n<td><code>Q2_K<\/code><\/td>\n<td>12.64GB<\/td>\n<td>5.0 tokens\/s<\/td>\n<td>8.9 chars\/s<\/td>\n<\/tr>\n<tr>\n<td>Qwopus3.8-27B-Flash<\/td>\n<td><code>Q3_K_M<\/code><\/td>\n<td>12.57GB<\/td>\n<td>0.9 tokens\/s<\/td>\n<td>1.7 chars\/s<\/td>\n<\/tr>\n<tr>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/15\/salience-27b-r6-gguf-2\/\">vectionlabs_Salience-27B-R6<\/a><\/td>\n<td><code>Q3_K_M<\/code><\/td>\n<td>12.48GB<\/td>\n<td>0.9 tokens\/s<\/td>\n<td>1.6 chars\/s<\/td>\n<\/tr>\n<tr>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/11\/pantheon-reasoning-26b-gguf\/\">Gryphe_Pantheon-Reasoning-26B-A4B-1.1-V2<\/a><\/td>\n<td><code>Q3_K_S<\/code><\/td>\n<td>12.26GB<\/td>\n<td>6.2 tokens\/s<\/td>\n<td>11.0 chars\/s<\/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 to Our Japanese Questions (Verbatim)<\/h3>\n<p>We ask every model the same five questions in Japanese under the conditions above (temperature 0, thinking turned off where the chat template allows it, up to 1024 tokens) and publish the answers unedited. We do not grade them; where a question has a single correct answer, we show it for comparison.<\/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<pre><code class=\"language-text\">\u307f\u3069\u308a\u5e02\u306f\u516c\u7acb\u4e2d\u592e\u56f3\u66f8\u9928\u3092\u7d041\u5e74\u534a\u30fb\u7dcf\u984d\u7d0412\u5104\u5186\u304b\u3051\u3066\u6539\u4fee\u3057\u3001\u95b2\u89a7\u5e2d\u3092\u5897\u3084\u3057\u3001\u4e00\u968e\u3092\u306b\u304e\u308f\u3044\u3001\u4e8c\u968e\u3092\u95b2\u89a7\u3001\u4e09\u968e\u3092\u5b66\u7fd2\u30b9\u30da\u30fc\u30b9\u306b\u5206\u3051\u308b\u306a\u3069\u3001\u5358\u306a\u308b\u8cb8\u51fa\u65bd\u8a2d\u304b\u3089\u4ea4\u6d41\u30fb\u751f\u6d3b\u652f\u63f4\u306e\u5834\u3078\u5909\u3048\u305f\u3002  \n\u8535\u66f8\u3067\u306f\u53e4\u3044\u8cc7\u6599\u3092\u6e1b\u3089\u3057\u3066\u7d75\u672c\u3084\u5b9f\u7528\u66f8\u3092\u5897\u3084\u3057\u3001\u96fb\u5b50\u66f8\u7c4d\u3001\u592a\u967d\u5149\u767a\u96fb\u3001\u30d0\u30ea\u30a2\u30d5\u30ea\u30fc\u8a2d\u5099\u3082\u5c0e\u5165\u3057\u305f\u304c\u3001\u99d0\u8eca\u5834\u4e0d\u8db3\u3084\u591c\u9593\u958b\u9928\u306b\u3088\u308b\u8077\u54e1\u8ca0\u62c5\u306a\u3069\u306e\u8ab2\u984c\u3082\u6b8b\u3063\u3066\u3044\u308b\u3002  \n\u5229\u7528\u8005\u306f\u6539\u4fee\u3092\u8a55\u4fa1\u3059\u308b\u4e00\u65b9\u3001\u5b66\u7fd2\u5e2d\u306e\u4e0d\u8db3\u3092\u6307\u6458\u3057\u3066\u304a\u308a\u3001\u5e02\u306f\u7a7a\u304d\u72b6\u6cc1\u306e\u8868\u793a\u5c0e\u5165\u3084\u6b8b\u308b\u56f3\u66f8\u9928\u306e\u6539\u4fee\u691c\u8a0e\u3092\u9032\u3081\u308b\u65b9\u91dd\u3060\u3002\n<\/code><\/pre>\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<pre><code class=\"language-text\">\u4ef6\u540d\uff1a\u6765\u9031\u706b\u66dc\u65e5\u306e\u6253\u3061\u5408\u308f\u305b\u65e5\u7a0b\u5909\u66f4\u306e\u304a\u9858\u3044\n\n\u3007\u3007\u69d8\n\n\u304a\u4e16\u8a71\u306b\u306a\u3063\u3066\u304a\u308a\u307e\u3059\u3002  \n\u3007\u3007\u3067\u3059\u3002\n\n\u5148\u65e5\u3054\u9023\u7d61\u3044\u305f\u3060\u3044\u305f\u3001\u6765\u9031\u706b\u66dc\u65e5\u306e\u6253\u3061\u5408\u308f\u305b\u306b\u3064\u304d\u307e\u3057\u3066\u3001  \n\u3054\u90fd\u5408\u3092\u8abf\u6574\u3044\u305f\u3060\u304d\u3001\u6765\u9031\u6c34\u66dc\u65e5\u306b\u5909\u66f4\u3057\u3066\u3044\u305f\u3060\u304f\u3053\u3068\u306f\u53ef\u80fd\u3067\u3057\u3087\u3046\u304b\u3002\n\n\u5909\u66f4\u306b\u3088\u308a\u3054\u4e0d\u4fbf\u3092\u304a\u304b\u3051\u3057\u7533\u3057\u8a33\u3054\u3056\u3044\u307e\u305b\u3093\u3002  \n\u3082\u3057\u6765\u9031\u6c34\u66dc\u65e5\u304c\u96e3\u3057\u3044\u5834\u5408\u306f\u3001\u3054\u90fd\u5408\u306e\u3088\u3044\u65e5\u6642\u3092\u3054\u6559\u793a\u3044\u305f\u3060\u3051\u307e\u3059\u3068\u5e78\u3044\u3067\u3059\u3002\n\n\u4f55\u5352\u3088\u308d\u3057\u304f\u304a\u9858\u3044\u3044\u305f\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<pre><code class=\"language-text\">\u308a\u3093\u3054\uff1a120\u5186 \u00d7 3\u500b = 360\u5186  \n\u307f\u304b\u3093\uff1a45\u5186 \u00d7 8\u500b = 360\u5186  \n\n\u5408\u8a08\uff1a360\u5186 + 360\u5186 = 720\u5186  \n\n1,000\u5186\u672d\u3067\u6255\u3063\u305f\u306e\u3067\u3001\u304a\u3064\u308a\u306f  \n1,000\u5186 \u2212 720\u5186 = **280\u5186**  \n\n\u3057\u305f\u304c\u3063\u3066\u3001\u304a\u3064\u308a\u306f **280\u5186** \u3067\u3059\u3002\n<\/code><\/pre>\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<pre><code class=\"language-text\">\u6d77\u306b\u9762\u3057\u3066\u3044\u306a\u3044\u770c\u306f\u3001\u4ee5\u4e0b\u306e8\u3064\u3067\u3059\u3002\n\n1. \u7fa4\u99ac\u770c  \n2. \u57fc\u7389\u770c  \n3. \u6803\u6728\u770c  \n4. \u5c71\u68a8\u770c  \n5. \u9577\u91ce\u770c  \n6. \u5c90\u961c\u770c  \n7. \u6ecb\u8cc0\u770c  \n8. \u4eac\u90fd\u5e9c  \n\n\u5408\u8a088\u90fd\u9053\u5e9c\u770c\u3067\u3059\u3002\n<\/code><\/pre>\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><em>(The model spent the whole token budget thinking and gave no answer.)<\/em><\/p>\n<p><!-- \/lmw:lab --><\/p>\n<p><!-- lmw:peers --><\/p>\n<h2>Recent Models in the Same Size Class<\/h2>\n<p><em>Models with <\/em><em>15\u201340B<\/em><em> parameters that Local Model Watch covered recently, listed by code from our article log for comparison. VRAM tiers are this site&#8217;s estimates; licenses are as stated on the model cards.<\/em><\/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>Parameters<\/th>\n<th>Smallest VRAM tier<\/th>\n<th>License<\/th>\n<th>Our article<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>elyza\/ELYZA-Thinking-1.0-llm-jp-4-32b-a3b<\/td>\n<td>32.1B<\/td>\n<td>80GB<\/td>\n<td>apache-2.0<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/10\/02\/elyza-thinking-1-0-32b-33b-2\/\">ELYZA Releases ELYZA-Thinking-1.0 32B\/33B Reasoning Models<\/a> (2026-10-02)<\/td>\n<\/tr>\n<tr>\n<td>XingChen-AGI\/Xing4.0-29B-A4B<\/td>\n<td>31.2B<\/td>\n<td>24GB<\/td>\n<td>apache-2.0<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/28\/xing4-0-29b-a4b-china-telecom\/\">Xing4.0-29B-A4B Text Generation Model: Our Test Answers, 24GB+ VRAM<\/a> (2026-09-28)<\/td>\n<\/tr>\n<tr>\n<td>Altworld\/Hemmingway-1<\/td>\n<td>26.9B<\/td>\n<td>12GB<\/td>\n<td>cc-by-nc-4.0<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/28\/hemmingway-1-open-27b-model-specialized-for-human-like-writing\/\">Hemmingway-1 Text Generation Model: 12GB+ VRAM, GGUF Builds<\/a> (2026-09-28)<\/td>\n<\/tr>\n<tr>\n<td>orcarouter\/OrcaSAQ-2-27B<\/td>\n<td>27.8B<\/td>\n<td>16GB<\/td>\n<td>apache-2.0<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/28\/orcasaq2-27b-qwen3-8-27b-quantized\/\">OrcaSAQ-2-27B Text Generation Model: Our Test Answers, 16GB+ VRAM<\/a> (2026-09-28)<\/td>\n<\/tr>\n<tr>\n<td>prism-ml\/Ternary-Bonsai-2-27B-gguf<\/td>\n<td>27.8B<\/td>\n<td>8GB<\/td>\n<td>apache-2.0<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/09\/18\/ternary-bonsai-2-27b-gguf\/\">Ternary-Bonsai-2-27B-gguf: Our Test Answers, 8GB+ VRAM<\/a> (2026-09-18)<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p><!-- \/lmw:peers --><\/p>\n<h2>How to Get It<\/h2>\n<p>Model weights are available as open source from Hugging Face and ModelScope.<\/p>\n<ul>\n<li>Hugging Face: <code>nex-agi\/Nex-N2.5-mini<\/code><\/li>\n<\/ul>\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>2026-09-18 (9 days after publication)<\/td>\n<td>bartowski<\/td>\n<td>GGUF (imatrix)<\/td>\n<td><a href=\"https:\/\/huggingface.co\/bartowski\/nex-agi_Nex-N2.5-mini-GGUF\">bartowski\/nex-agi_Nex-N2.5-mini-GGUF<\/a><\/td>\n<td>Q2_K 15.5GB (fits in 16GB VRAM)<\/td>\n<\/tr>\n<tr>\n<td>2026-09-18 (9 days after publication)<\/td>\n<td>mradermacher<\/td>\n<td>GGUF<\/td>\n<td><a href=\"https:\/\/huggingface.co\/mradermacher\/Nex-N2.5-mini-GGUF\">mradermacher\/Nex-N2.5-mini-GGUF<\/a><\/td>\n<td>Q2_K 14.5GB (fits in 16GB VRAM)<\/td>\n<\/tr>\n<tr>\n<td>2026-09-18 (9 days after publication)<\/td>\n<td>mlx-community<\/td>\n<td>MLX<\/td>\n<td><a href=\"https:\/\/huggingface.co\/mlx-community\/Nex-N2.5-mini-OptiQ-4bit\">mlx-community\/Nex-N2.5-mini-OptiQ-4bit<\/a><\/td>\n<td>MLX 4bit 24.8GB (fits in 32GB VRAM)<\/td>\n<\/tr>\n<tr>\n<td>2026-09-18 (9 days after publication)<\/td>\n<td>mlx-community<\/td>\n<td>MLX<\/td>\n<td><a href=\"https:\/\/huggingface.co\/mlx-community\/Nex-N2.5-mini-oQ4\">mlx-community\/Nex-N2.5-mini-oQ4<\/a><\/td>\n<td>MLX 4bit 22.8GB (fits in 24GB 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 bartowski\/nex-agi_Nex-N2.5-mini-GGUF: IQ2_XXS 10.1GB \/ IQ2_XS 10.5GB \/ IQ2_S 11.2GB \/ IQ2_M 12.3GB \/ Q2_K 12.9GB \/ IQ3_XXS 14.5GB \/ IQ3_XS 15.2GB \/ Q3_K_S 15.2GB \/ Q3_K_M 16.2GB \/ Q3_K_L 17.0GB \/ IQ3_M 17.7GB \/ IQ4_XS 18.5GB \/ Q4_0 18.6GB \/ Q4_K_S 19.5GB \/ Q4_1 20.5GB \/ IQ4_NL 20.8GB \/ Q4_K_M 20.8GB \/ Q4_K_L 22.5GB \/ Q5_K_S 23.4GB \/ Q5_K_M 25.1GB \/ Q6_K 27.4GB \/ Q6_K_L 30.1GB \/ Q8_0 34.4GB \/ BF16 64.6GB<\/li>\n<li>Available builds in mradermacher\/Nex-N2.5-mini-GGUF: Q2_K 12.1GB \/ Q3_K_S 14.1GB \/ Q3_K_M 15.6GB \/ Q3_K_L 16.9GB \/ IQ4_XS 17.6GB \/ Q4_K_S 18.5GB \/ Q4_K_M 19.7GB \/ Q5_K_S 22.3GB \/ Q5_K_M 23.0GB \/ Q6_K 26.6GB \/ Q8_0 34.4GB<\/li>\n<li>Available builds in mlx-community\/Nex-N2.5-mini-OptiQ-4bit: MLX 4bit 20.6GB<\/li>\n<li>Available builds in mlx-community\/Nex-N2.5-mini-oQ4: MLX 4bit 19.0GB<\/li>\n<\/ul>\n<p>In addition, 33 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\/09\/nex-agi-announces-nex-n25-pro-agent-model\/\">Nex-N2.5-Pro Long-Horizon Agent Model: ~98GB Memory, GGUF Builds<\/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 16GB tier) \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/vram-guide-16gb-en\/\">Other models that run on a 16GB GPU<\/a><\/li>\n<li><strong>Explore the same model family<\/strong> \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/model-nex-agi-nex-n2-5-en\/\">Nex-N2.5 family overview (2 articles, 6 converted builds)<\/a><\/li>\n<li><strong>Engines that run this model<\/strong> \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/engine-llama-cpp-en\/\">llama.cpp<\/a> \/ <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/engine-ollama-en\/\">Ollama<\/a> \/ <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/engine-vllm-en\/\">vLLM<\/a><\/li>\n<li><strong>What Q2_K, Q4_K_S, Q5_K_M 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\/format-mlx-en\/\">MLX 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-nex-agi-en\/\">Nex-AGI: 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\/nex-agi\/Nex-N2.5-mini\">https:\/\/huggingface.co\/nex-agi\/Nex-N2.5-mini<\/a><\/li>\n<\/ul>\n<p><!-- lmw:updates --><\/p>\n<h2>Update History<\/h2>\n<ul>\n<li>2026-09-18: Added converted builds to \u201cQuantized and Converted Variants\u201d: bartowski\/nex-agi_Nex-N2.5-mini-GGUF, mradermacher\/Nex-N2.5-mini-GGUF, mlx-community\/Nex-N2.5-mini-OptiQ-4bit, mlx-community\/Nex-N2.5-mini-oQ4<\/li>\n<li>2026-09-18: Updated the hardware requirements table with the actual file sizes of bartowski\/nex-agi_Nex-N2.5-mini-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>2026-09-25: Changed the title to show what the article covers (VRAM requirements, file list, etc.).<\/li>\n<li>2026-09-28: 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>2026-09-30: Changed the title to show what the article covers (VRAM requirements, file list, etc.).<\/li>\n<\/ul>\n<p><!-- \/lmw:updates --><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Discover the specs, performance, and hardware requirements for Nex-N2.5-mini, an open-weight multimodal model by Nex-AGI for long tasks.<\/p>\n","protected":false},"author":1,"featured_media":7645,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[310],"tags":[165,549,551,377,896,117],"class_list":["post-354","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-new-models","tag-moe-en","tag-nex-agi-en","tag-nex-n2-5-mini-en","tag--en"],"lang":"en","translations":{"en":354,"ja":352},"_links":{"self":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/posts\/354","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=354"}],"version-history":[{"count":81,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/posts\/354\/revisions"}],"predecessor-version":[{"id":14815,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/posts\/354\/revisions\/14815"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/media\/7645"}],"wp:attachment":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/media?parent=354"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/categories?post=354"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/tags?post=354"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}