{"id":11478,"date":"2026-11-17T11:00:01","date_gmt":"2026-11-17T02:00:01","guid":{"rendered":"https:\/\/localmodelwatch.tsuchitsuchi.com\/?p=11478"},"modified":"2026-11-17T11:09:22","modified_gmt":"2026-11-17T02:09:22","slug":"liquidai-lfm2-5-24b-a2b-fp8-released","status":"publish","type":"post","link":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/11\/17\/liquidai-lfm2-5-24b-a2b-fp8-released\/","title":{"rendered":"LFM2.5-24B-A2B-FP8 Text Generation Model: 32GB+ 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\/LiquidAI\/LFM2.5-24B-A2B-FP8\">LiquidAI\/LFM2.5-24B-A2B-FP8<\/a><\/td>\n<\/tr>\n<tr>\n<td>Family guide<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/model-liquidai-lfm2-5-en\/\">LFM2.5 guide (11 articles)<\/a><\/td>\n<\/tr>\n<tr>\n<td>Publisher guide<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/publisher-liquidai-en\/\">Liquid AI: models and licenses<\/a><\/td>\n<\/tr>\n<tr>\n<td>Published<\/td>\n<td>2026-10-13<\/td>\n<\/tr>\n<tr>\n<td>License<\/td>\n<td>lfm1.0<\/td>\n<\/tr>\n<tr>\n<td>Formats<\/td>\n<td>safetensors<\/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<h2>Overview<\/h2>\n<p>Liquid AI has released <code>LiquidAI\/LFM2.5-24B-A2B-FP8<\/code>, an FP8 quantized version of their inference model <code>LFM2.5-24B-A2B<\/code>.<\/p>\n<p>This model is a checkpoint tailored for GPU inference, where the weights of the linear layers are quantized to FP8 in 128&#215;128 block units, combined with dynamic activation quantization. Meanwhile, the embedding layers, Mixture of Experts (MoE) routing gates, and convolutional layer weights retain their original precision.<\/p>\n<p>It aims to achieve high-throughput serving with reduced memory consumption when using inference engines such as vLLM and SGLang.<\/p>\n<h2>Specifications<\/h2>\n<p>According to the original model&#8217;s model card, the specifications are as follows:<\/p>\n<ul>\n<li>Total parameters: 24.0B<\/li>\n<li>Active parameters: 2.45B (noted as 2B in the model overview)<\/li>\n<li>Architecture: MoE (Lfm2MoeForCausalLM, 40 layers in total consisting of 30 double-gated short convolution blocks and 10 GQA blocks)<\/li>\n<li>Expert configuration: 64 experts per MoE block (top-4 routing)<\/li>\n<li>Context length: 131,072 tokens<\/li>\n<li>Vocabulary size: 128,000<\/li>\n<\/ul>\n<h2>Performance<\/h2>\n<p>Since the newly published repository itself does not contain a benchmark table, the measurement results published in the model card for the original <code>LFM2.5-24B-A2B<\/code> model are presented below. Note that because this model has undergone FP8 quantization, slight performance variations may occur compared to the original precision.<\/p>\n<p>Comparison results with major open-weight MoE models provided by the publisher are shown in the table below.<\/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>Benchmark<\/th>\n<th>LFM2.5-24B-A2B<\/th>\n<th>Qwen3.6-35B-A3B<\/th>\n<th>Gemma-4-26B-A4B-it<\/th>\n<th>Nemotron-3.5-Lightning (30B-A3B)<\/th>\n<\/tr>\n<\/thead>\n<tbody>\n<tr>\n<td>AutomationBench<\/td>\n<td>27.69<\/td>\n<td>8.72<\/td>\n<td>5.46<\/td>\n<td>2.61<\/td>\n<\/tr>\n<tr>\n<td>BFCLv4<\/td>\n<td>65.01<\/td>\n<td>65.00<\/td>\n<td>57.09<\/td>\n<td>53.78<\/td>\n<\/tr>\n<tr>\n<td>ToolSandbox<\/td>\n<td>80.71<\/td>\n<td>75.02<\/td>\n<td>72.09<\/td>\n<td>75.13<\/td>\n<\/tr>\n<tr>\n<td>LiveClaw<\/td>\n<td>48.09<\/td>\n<td>38.19<\/td>\n<td>41.25<\/td>\n<td>37.72<\/td>\n<\/tr>\n<tr>\n<td>PinchBench<\/td>\n<td>83.18<\/td>\n<td>89.93<\/td>\n<td>77.12<\/td>\n<td>84.20<\/td>\n<\/tr>\n<tr>\n<td>\u03c4\u00b3-Bench Banking<\/td>\n<td>8.04<\/td>\n<td>7.63<\/td>\n<td>4.54<\/td>\n<td>7.01<\/td>\n<\/tr>\n<tr>\n<td>WebArena Verified<\/td>\n<td>26.75<\/td>\n<td>37.62<\/td>\n<td>25.62<\/td>\n<td>28.69<\/td>\n<\/tr>\n<tr>\n<td>GPQA Diamond<\/td>\n<td>74.42<\/td>\n<td>83.69<\/td>\n<td>74.90<\/td>\n<td>72.42<\/td>\n<\/tr>\n<tr>\n<td>AA-Omniscience<\/td>\n<td>-24.90<\/td>\n<td>-24.35<\/td>\n<td>-59.39<\/td>\n<td>-3.58<\/td>\n<\/tr>\n<tr>\n<td>HLE<\/td>\n<td>7.40<\/td>\n<td>7.69<\/td>\n<td>8.57<\/td>\n<td>5.65<\/td>\n<\/tr>\n<tr>\n<td>IFEval<\/td>\n<td>96.75<\/td>\n<td>93.06<\/td>\n<td>92.00<\/td>\n<td>94.19<\/td>\n<\/tr>\n<tr>\n<td>IFBench<\/td>\n<td>74.69<\/td>\n<td>64.21<\/td>\n<td>47.44<\/td>\n<td>71.67<\/td>\n<\/tr>\n<tr>\n<td>Multi-IF<\/td>\n<td>84.38<\/td>\n<td>81.53<\/td>\n<td>82.19<\/td>\n<td>82.34<\/td>\n<\/tr>\n<tr>\n<td>MATH-500<\/td>\n<td>96.04<\/td>\n<td>96.73<\/td>\n<td>95.66<\/td>\n<td>96.40<\/td>\n<\/tr>\n<tr>\n<td>SWE-Bench Verified (OpenCode)<\/td>\n<td>51.00<\/td>\n<td>66.20<\/td>\n<td>25.80<\/td>\n<td>53.40<\/td>\n<\/tr>\n<tr>\n<td>SciCode<\/td>\n<td>32.71<\/td>\n<td>37.35<\/td>\n<td>35.42<\/td>\n<td>27.78<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p>From these measurement results, the original model scores 96.75 on IFEval, which measures adherence to formal instructions, demonstrating the highest instruction-following performance among the compared models. It also achieves high scores, significantly pulling away from other models, in agentic benchmarks such as AutomationBench (27.69) and LiveClaw (48.09).<\/p>\n<p>While it maintains a high standard of 96.04 on MATH-500 for solving high school to competition-level mathematics, it falls slightly behind Qwen3.6-35B-A3B (96.73) and Nemotron-3.5-Lightning (96.40).<\/p>\n<p>On the other hand, it stops at 7.40 on HLE, a collection of extremely difficult expert-created questions, falling below Gemma-4-26B-A4B-it (8.57) and Qwen3.6-35B-A3B (7.69). Furthermore, it scores 74.42 on GPQA Diamond, which covers PhD-level physics, chemistry, and biology questions, lagging far behind Qwen3.6-35B-A3B&#8217;s 83.69. It is also confirmed that there are areas where competitors outperform it, such as in SWE-Bench Verified (51.00 vs 66.20), which measures problem-solving capabilities requiring advanced knowledge and task-fixing skills in real repositories.<\/p>\n<h2>Strengths and Use Cases<\/h2>\n<p>As this is an FP8 quantized version, its strengths and functional characteristics conform to those of the original <code>LFM2.5-24B-A2B<\/code> model.<\/p>\n<p>The original model is designed with a focus on agentic workloads that run fast and privately on devices or local environments. It is a reasoning model that always thinks before outputting an answer, exhibiting behavior where a <code>&lt;think&gt;<\/code> tag is automatically inserted inside the chat template at the start of the assistant&#8217;s response. Because its active parameter count is kept low at 2.45B (noted as 2B in the overview), its strength lies in generating thinking tokens at a low cost while keeping computation per token low.<\/p>\n<p>It is particularly proficient in agentic tasks involving Tool Use \/ Function Calling. By providing tool definitions in JSON format within the system prompt, it can coordinate with external tools through four steps (definition, call, execution result input, final answer). By default, it outputs Python-style function calls enclosed in special tokens <code>&lt;|tool_call_start|&gt;<\/code> and <code>&lt;|tool_call_end|&gt;<\/code>, but it can also be made to output in JSON format through instructions in the system prompt. Additionally, during the post-training phase, it has undergone reinforcement learning and distillation within representative agent harnesses such as OpenClaw, Hermes Agent, Pi, and OpenCode, adapting to the prompt structures and operation patterns of these tools.<\/p>\n<p>It supports a context length of 131,072 tokens and multilingual capabilities including Japanese, English, Arabic, Chinese, French, German, Italian, Korean, Portuguese, Spanish, Vietnamese, Thai, Indonesian, Hindi, Russian, and Polish. The vocabulary size has also been expanded to 128,000, improving processing efficiency for languages containing non-Latin characters.<\/p>\n<p>Meanwhile, the model card does not recommend letting this model handle knowledge-intensive tasks entirely on its own. Rather than comprehensively covering facts independently, it is recommended to leverage the model&#8217;s strong tool-calling capabilities in combination with retrieval systems (RAG) or web search.<\/p>\n<p><!-- lmw:hardware --><\/p>\n<h2>Hardware Requirements<\/h2>\n<p><strong>Estimated requirements (calculated by Local Model Watch)<\/strong> \u2014 24.0B 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>32GB (RTX 5090, etc.)<\/td>\n<td>FP8<\/td>\n<td>23.0GB<\/td>\n<td>27.7GB<\/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 2026-11-17): 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. 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>Not usable in Ollama, LM Studio and llama.cpp yet \u2014 we have found no GGUF build.<\/strong><\/p>\n<p>The publisher ships safetensors only. However, llama.cpp&#8217;s registry does list this architecture, so <strong>conversion to GGUF is possible<\/strong> and the model will run once someone publishes a converted build. Today it can be run with transformers or vLLM, using the memory figures in the table above.<\/p>\n<p><strong>License \u2014 <code>lfm1.0<\/code> (Commercial use allowed with conditions):<\/strong> LiquidAI&#8217;s LFM Open License v1.0. Commercial use is allowed, except by legal entities with annual revenue of $10 million or more (Section 5). Redistribution requires the license text and notice of changes.<\/p>\n<p><strong>Compression:<\/strong> the FP8 build measures 8.24 bits per weight \u2014 about 52% 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: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>IFM\/Jais-3-32B-GGUF<\/td>\n<td>34.8B<\/td>\n<td>24GB<\/td>\n<td>apache-2.0<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/11\/17\/jais-3-32b-gguf-released\/\">Jais-3-32B-GGUF Arabic and English 32B Dense Model: 24GB+ VRAM<\/a> (2026-11-17)<\/td>\n<\/tr>\n<tr>\n<td>IFM\/Jais-3-MoVA-36B-A4B-GGUF<\/td>\n<td>37.4B<\/td>\n<td>32GB<\/td>\n<td>apache-2.0<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/11\/17\/jais-3-mova-36b-a4b-gguf\/\">Jais-3-MoVA-36B-A4B-GGUF Text Generation Model: 32GB+ VRAM<\/a> (2026-11-17)<\/td>\n<\/tr>\n<tr>\n<td>tencent\/Hy-Memory-30B-A3B<\/td>\n<td>30.1B<\/td>\n<td>12GB<\/td>\n<td>apache-2.0<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/11\/17\/hy-memory-30b-a3b-2\/\">Hy-Memory-30B-A3B Text Generation Model: 12GB+ VRAM, GGUF Builds<\/a> (2026-11-17)<\/td>\n<\/tr>\n<tr>\n<td>KRAFTON\/Raon-1-mini-preview<\/td>\n<td>28.8B<\/td>\n<td>24GB<\/td>\n<td>apache-2.0<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/11\/17\/krafton-ai-raon-1-mini-preview\/\">Raon-1-mini-preview Reasoning-Enabled Language Model: 24GB+ VRAM<\/a> (2026-11-17)<\/td>\n<\/tr>\n<tr>\n<td>ggml-org\/Flint-GGUF<\/td>\n<td>39.8B<\/td>\n<td>32GB<\/td>\n<td>apache-2.0<\/td>\n<td><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/11\/17\/ggml-org-flint-gguf\/\">Flint-GGUF Coding MoE Language Model: 32GB+ VRAM<\/a> (2026-11-17)<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n<p><!-- \/lmw:peers --><\/p>\n<h2>How to Get It<\/h2>\n<p>This model is distributed in safetensors format on the Hugging Face repository <code>LiquidAI\/LFM2.5-24B-A2B-FP8<\/code>. It is not a gated model requiring prior access requests or license agreements for use, but the &#8220;lfm1.0&#8221; license applies.<\/p>\n<p>This checkpoint is optimized to deliver high throughput while reducing memory usage during GPU inference by applying FP8 block quantization to linear layer weights combined with dynamic activation quantization. SGLang and vLLM are cited as supported engines, and when using SGLang, you can launch the inference server with the following command:<\/p>\n<pre><code class=\"language-bash\">python -m sglang.launch_server --model-path LiquidAI\/LFM2.5-24B-A2B-FP8\n<\/code><\/pre>\n<p><!-- lmw:related --><\/p>\n<h2>Related Articles<\/h2>\n<ul>\n<li><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/11\/17\/liquidai-lfm2-5-24b-a2b-mlx-8bit\/\">LFM2.5-24B-A2B-MLX-8bit Text Generation Model: 32GB+ VRAM<\/a><\/li>\n<li><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/11\/17\/liquid-ai-lfm25-24b-a2b-dspark\/\">LFM2.5-24B-A2B-DSpark Speculative Decoding Draft Model: 4GB+ VRAM<\/a><\/li>\n<li><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/11\/17\/liquid-ai-lfm2-5-24b-a2b-mlx-6bit\/\">LFM2.5-24B-A2B-MLX-6bit Text Generation Model: 24GB+ VRAM<\/a><\/li>\n<li><a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/2026\/11\/17\/liquidai-lfm25-24b-a2b\/\">LFM2.5-24B-A2B Agent-Optimized MoE Reasoning 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 needs at least 32GB) \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/vram-guide-en\/\">VRAM quick reference<\/a><\/li>\n<li><strong>Explore the same model family<\/strong> \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/model-liquidai-lfm2-5-en\/\">LFM2.5 family overview (11 articles, 17 converted builds)<\/a><\/li>\n<li><strong>Engines that run this model<\/strong> \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/engine-vllm-en\/\">vLLM<\/a><\/li>\n<li><strong>What FP8 mean and where to get this model<\/strong> \u2192 <a href=\"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/format-fp8-en\/\">FP8 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-liquidai-en\/\">Liquid AI: 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\/LiquidAI\/LFM2.5-24B-A2B-FP8\">https:\/\/huggingface.co\/LiquidAI\/LFM2.5-24B-A2B-FP8<\/a><\/li>\n<li><a href=\"https:\/\/huggingface.co\/LiquidAI\/LFM2.5-24B-A2B\">https:\/\/huggingface.co\/LiquidAI\/LFM2.5-24B-A2B<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>Liquid AI has released LiquidAI\/LFM2.5-24B-A2B-FP8, an FP8 quantized version of their inference model designed for efficient GPU serving.<\/p>\n","protected":false},"author":1,"featured_media":11477,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[310],"tags":[1133,2959,3272,2278,165,169,1547,592,1555],"class_list":["post-11478","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-new-models","tag-fp8-en","tag-lfm2-5-en","tag-lfm2-5-24b-a2b-en","tag-liquidai-en","tag-moe-en","tag-sglang-en","tag-verified","tag-vllm-en","tag--en"],"lang":"en","translations":{"en":11478,"ja":11476},"_links":{"self":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/posts\/11478","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=11478"}],"version-history":[{"count":2,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/posts\/11478\/revisions"}],"predecessor-version":[{"id":17761,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/posts\/11478\/revisions\/17761"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/media\/11477"}],"wp:attachment":[{"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/media?parent=11478"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/categories?post=11478"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/localmodelwatch.tsuchitsuchi.com\/en\/wp-json\/wp\/v2\/tags?post=11478"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}