AI Index

Open-weight models that run on CPU only · 96 GB

  • 96 GB DDR5 dual channel · 89.6 GB/s
  • Usable 90.0 GB
FitModelParameters / activePrecisionMemoryEst. speedLongest contextRuns as
Runs wellLing-3.0-flash Ant Group127.5B / 5.1BQ4_K_Mmaker’s file75.7 / 90.0 GB13–19 tok/s250KSystem memory
Runs wellLing-3.0-flash-VL Ant Group124.8B / 5.5BQ4_069.0 / 90.0 GB13–19 tok/s128KSystem memory
Runs wellgpt-oss-120b OpenAI116.8B / 5.1BMXFP4 (native)maker’s file64.1 / 90.0 GB14–21 tok/s128KSystem memory
Runs wellQwen3-Next-80B-A3B-Instruct Alibaba Qwen81.3B / 3BFP8maker’s file80.4 / 90.0 GB14–21 tok/s256KSystem memory
Runs wellQwen3-Next-80B-A3B-Thinking Alibaba Qwen81.3B / 3BFP8maker’s file80.4 / 90.0 GB14–21 tok/s256KSystem memory
Runs wellQwen3-Coder-Next Alibaba Qwen79.7B / 3BFP8maker’s file78.8 / 90.0 GB14–21 tok/s256KSystem memory
Runs wellQwen3.5-35B-A3B Alibaba Qwen36B / 3BQ8_037.5 / 90.0 GB13–20 tok/s256KSystem memory
Runs wellQwen3.6-35B-A3B Alibaba Qwen36B / 3BQ8_037.5 / 90.0 GB13–20 tok/s256KSystem memory
Runs wellQwen3-Omni-30B-A3B-Captioner Alibaba Qwen31.7BQ4_K_M19.6 / 90.0 GB13–19 tok/s64KSystem memory
Runs wellGLM-4.7-Flash Zhipu AI31.2BQ4_017.6 / 90.0 GB13–20 tok/s195KSystem memory
Runs wellQwen3-VL-30B-A3B-Instruct Alibaba Qwen31.1BQ5_K_M22.4 / 90.0 GB13–19 tok/s128KSystem memory
Runs wellQwen3-VL-30B-A3B-Thinking Alibaba Qwen31.1BQ5_K_M22.4 / 90.0 GB13–19 tok/s128KSystem memory
Runs wellQwen3-30B-A3B-Instruct-2507 Alibaba Qwen30.5B / 3.3BQ6_K25.3 / 90.0 GB13–19 tok/s256KSystem memory
Runs wellQwen3-30B-A3B-Thinking-2507 Alibaba Qwen30.5B / 3.3BQ6_K25.3 / 90.0 GB13–19 tok/s256KSystem memory
Runs wellQwen3-30B-A3B Alibaba Qwen30.5B / 3.3BQ6_Kmaker’s file25.3 / 90.0 GB13–19 tok/s128KSystem memory
Runs wellQwen3-Coder-30B-A3B-Instruct Alibaba Qwen30.5B / 3.3BQ6_K25.3 / 90.0 GB13–19 tok/s256KSystem memory
Runs wellHy-Memory-30B-A3B Tencent30.1B / 3BQ6_Kmaker’s file24.9 / 90.0 GB14–21 tok/s63KSystem memory
Runs wellGemma 4 26B A4B IT Google25.8B / 3.8BQ5_K_M18.5 / 90.0 GB14–21 tok/s250KSystem memory
Runs wellERNIE-4.5-21B-A3B-Thinking Baidu21.8B / 3BFP821.8 / 90.0 GB13–19 tok/s128KSystem memory
Runs wellGPT OSS 20B OpenAI20.9B / 3.6BMXFP4 (native)maker’s file13.6 / 90.0 GB17–26 tok/s128KSystem memory
Runs wellLing-3.0-tiny Ant Group7.9B / 1.3BBF16 (native)maker’s file15.6 / 90.0 GB16–24 tok/s128KSystem memory
Runs wellGemma 4 E2B Google5.1BQ4_K_M3.5 / 90.0 GB14–21 tok/s125KSystem memory
Runs wellQwen3.5-4B Alibaba Qwen4.7BQ5_K_M3.8 / 90.0 GB13–19 tok/s256KSystem memory
Runs wellQwen3-VL-4B-Instruct Alibaba Qwen4.4BIQ4_XS3.8 / 90.0 GB13–19 tok/s250KSystem memory
Runs wellQwen3-4B-Instruct-2507 Alibaba Qwen4BQ4_03.7 / 90.0 GB13–19 tok/s256KSystem memory
Runs wellQwen3-4B Alibaba Qwen4BQ4_03.7 / 90.0 GB13–19 tok/s128KSystem memory
Runs wellDeepSeek-OCR DeepSeek3.3BBF16 (native)maker’s file7.2 / 90.0 GB16–24 tok/s8KSystem memory
Runs wellGranite-4.0-H-Micro IBM3.2BQ8_0maker’s file3.7 / 90.0 GB13–19 tok/s128KSystem memory
Runs wellQwen3-1.7B Alibaba Qwen2BQ8_0maker’s file3.1 / 90.0 GB16–24 tok/s40KSystem memory
Runs wellDeepSeek-R1-Distill-Qwen-1.5B DeepSeek1.8BQ8_02.5 / 90.0 GB21–32 tok/s125KSystem memory
Runs wellGLM-OCR Zhipu AI1.3BBF16 (native)maker’s file3.5 / 90.0 GB14–21 tok/s128KSystem memory
Runs wellQwen3-0.6B Alibaba Qwen0.8BBF16 (native)maker’s file2.8 / 90.0 GB18–28 tok/s40KSystem memory
Runs wellEmbeddingGemma 2 Google0.7BBF16 (native)maker’s file2.0 / 90.0 GB28–43 tok/s256KSystem memory
Runs wellQwen3-Embedding-0.6B Alibaba Qwen0.6BBF16 (native)maker’s file2.5 / 90.0 GB21–32 tok/s32KSystem memory
Runs wellbge-reranker-v2-m3 BAAI0.6BFP32 (native)maker’s file3.4 / 90.0 GB15–22 tok/s8KSystem memory
Runs wellbge-m3-unsupervised BAAI0.6BFP32 (native)maker’s file3.4 / 90.0 GB15–22 tok/s8KSystem memory
Runs wellbge-large-en-v1.5 BAAI0.3BFP32 (native)maker’s file1.8 / 90.0 GB32–48 tok/s512System memory
Runs wellbge-reranker-base BAAI0.3BFP32 (native)maker’s file1.6 / 90.0 GB40–59 tok/s514System memory
Runs wellbge-base-en-v1.5 BAAI0.1BFP32 (native)maker’s file0.9 / 90.0 GB98–147 tok/s512System memory
Runs wellbge-small-en-v1.5 BAAI0BFP32 (native)maker’s file0.6 / 90.0 GB313–470 tok/s512System memory
Runs, slowlyMing-3.0-audio-flash Ant Group128.2B / 6.5BQ4_K_M76.5 / 90.0 GB10–15 tok/s128KSystem memory
Runs, slowlyQwen3.5-122B-A10B Alibaba Qwen125.1B / 10BQ4_K_M74.5 / 90.0 GB7–11 tok/s256KSystem memory
Runs, slowlyGLM-4.5-Air Zhipu AI110.5BQ5_K_M78.3 / 90.0 GB3–5 tok/s73KSystem memory
Runs, slowlyGLM-4.5V Zhipu AI107.7B / 12BQ5_K_M76.4 / 90.0 GB4–7 tok/s63KSystem memory
Runs, slowlyGLM-4.6V Zhipu AI107.7BQ5_K_M76.4 / 90.0 GB4–6 tok/s84KSystem memory
Runs, slowlyLing-2.6-flash Ant Group107.5B / 7.4BQ5_K_M75.0 / 90.0 GB8–12 tok/s128KSystem memory
Runs, slowlyQwen2.5-VL-72B-Instruct Alibaba Qwen73.4BQ8_078.8 / 90.0 GB1–1 tok/s44KSystem memory
Runs, slowlyQwen2-VL-72B-Instruct Alibaba Qwen73.4BQ8_078.8 / 90.0 GB1–1 tok/s32KSystem memory
Runs, slowlyQwen2.5-72B-Instruct Alibaba Qwen72.7BQ8_0maker’s file78.3 / 90.0 GB1–1 tok/s32KSystem memory
Runs, slowlyQianfan-VL-70B Baidu71.8BQ8_077.1 / 90.0 GB1–1 tok/s32KSystem memory
Runs, slowlyDeepSeek-R1-Distill-Llama-70B DeepSeek70.6BQ8_075.8 / 90.0 GB1–1 tok/s53KSystem memory
Runs, slowlyQwen2.5-VL-32B-Instruct Alibaba Qwen33.5BBF16 (native)maker’s file68.8 / 90.0 GB1–1 tok/s93KSystem memory
Runs, slowlyQwen3-VL-32B-Instruct Alibaba Qwen33.4BBF16 (native)maker’s file67.2 / 90.0 GB1–1 tok/s99KSystem memory
Runs, slowlyQwen3-VL-32B-Thinking Alibaba Qwen33.4BBF16 (native)maker’s file67.2 / 90.0 GB1–1 tok/s99KSystem memory
Runs, slowlyQwen2.5-32B-Instruct Alibaba Qwen32.8BBF16 (native)maker’s file66.1 / 90.0 GB1–1 tok/s32KSystem memory
Runs, slowlyQwen2.5-Coder-32B-Instruct Alibaba Qwen32.8BBF16 (native)maker’s file66.1 / 90.0 GB1–1 tok/s32KSystem memory
Runs, slowlyQwQ-32B Alibaba Qwen32.8BBF16 (native)maker’s file66.1 / 90.0 GB1–1 tok/s40KSystem memory
Runs, slowlyDeepSeek-R1-Distill-Qwen-32B DeepSeek32.8BBF16 (native)maker’s file66.1 / 90.0 GB1–1 tok/s32KSystem memory
Runs, slowlyQwen3-32B Alibaba Qwen32.8BBF16 (native)maker’s file66.1 / 90.0 GB1–1 tok/s128KSystem memory
Runs, slowlyGemma 4 31B IT Google31.3BBF16 (native)maker’s file63.2 / 90.0 GB1–1 tok/s179KSystem memory
Runs, slowlyGranite Switch 4.1 30B IBM30BBF16 (native)maker’s file60.7 / 90.0 GB1–1 tok/s122KSystem memory
Runs, slowlyQwen3.5-27B Alibaba Qwen27.8BBF16 (native)maker’s file54.8 / 90.0 GB1–1 tok/s256KSystem memory
Runs, slowlyQwen3.6-27B Alibaba Qwen27.8BBF16 (native)maker’s file54.8 / 90.0 GB1–1 tok/s256KSystem memory
Runs, slowlyQwen3.8-27B Alibaba Qwen27.8BBF16 (native)maker’s file54.8 / 90.0 GB1–1 tok/s256KSystem memory
Runs, slowlyMagistral Small 1.2 Mistral AI24BQ8_0maker’s file25.8 / 90.0 GB2–3 tok/s125KSystem memory
Runs, slowlyDevstral Small 1.0 Mistral AI23.6BQ8_025.7 / 90.0 GB2–3 tok/s125KSystem memory
Runs, slowlyDevstral Small 1.1 Mistral AI23.6BQ8_025.7 / 90.0 GB2–3 tok/s125KSystem memory
Runs, slowlyMagistral Small 1.0 Mistral AI23.6BQ8_025.7 / 90.0 GB2–3 tok/s39KSystem memory
Runs, slowlyMagistral Small 1.1 Mistral AI23.6BQ8_0maker’s file25.8 / 90.0 GB2–3 tok/s39KSystem memory
Runs, slowlyQwen2.5-14B-Instruct-1M Alibaba Qwen14.8BBF16 (native)maker’s file30.4 / 90.0 GB1–2 tok/s326KSystem memory
Runs, slowlyQwen2.5-14B-Instruct Alibaba Qwen14.8BBF16 (native)maker’s file30.4 / 90.0 GB1–2 tok/s32KSystem memory
Runs, slowlyDeepSeek-R1-Distill-Qwen-14B DeepSeek14.8BBF16 (native)maker’s file30.4 / 90.0 GB1–2 tok/s32KSystem memory
Runs, slowlyQwen3-14B Alibaba Qwen14.8BBF16 (native)maker’s file30.1 / 90.0 GB1–2 tok/s128KSystem memory
Runs, slowlyGemma 3 12B IT Google12.2BBF16 (native)maker’s file26.6 / 90.0 GB2–2 tok/s128KSystem memory
Runs, slowlyQwen2.5-Omni-7B Alibaba Qwen10.7BBF16 (native)maker’s file22.3 / 90.0 GB2–3 tok/s32KSystem memory
Runs, slowlyGLM-4.6V-Flash Zhipu AI10.3BBF16 (native)maker’s file20.4 / 90.0 GB2–3 tok/s125KSystem memory
Runs, slowlyQwen3.5-9B Alibaba Qwen9.7BBF16 (native)maker’s file19.1 / 90.0 GB2–3 tok/s256KSystem memory
Runs, slowlyQianfan-VL-8B Baidu8.8BBF16 (native)maker’s file18.2 / 90.0 GB2–4 tok/s32KSystem memory
Runs, slowlyGranite Switch 4.1 8B IBM8.8BBF16 (native)maker’s file18.5 / 90.0 GB2–4 tok/s128KSystem memory
Runs, slowlyQwen3-VL-8B-Instruct Alibaba Qwen8.8BBF16 (native)maker’s file18.3 / 90.0 GB2–4 tok/s128KSystem memory
Runs, slowlyQwen3-VL-8B-Thinking Alibaba Qwen8.8BBF16 (native)maker’s file18.3 / 90.0 GB2–4 tok/s128KSystem memory
Runs, slowlypplx-embed-v2-context-9b Perplexity8.4BFP32 (native)maker’s file33.1 / 90.0 GB1–2 tok/s256KSystem memory
Runs, slowlyQwen2.5-VL-7B-Instruct Alibaba Qwen8.3BBF16 (native)maker’s file16.7 / 90.0 GB3–4 tok/s125KSystem memory
Runs, slowlyQwen3-8B Alibaba Qwen8.2BBF16 (native)maker’s file17.1 / 90.0 GB3–4 tok/s128KSystem memory
Runs, slowlyDeepSeek-R1-Distill-Llama-8B DeepSeek8BBF16 (native)maker’s file16.7 / 90.0 GB3–4 tok/s125KSystem memory
Runs, slowlyGemma 4 E4B Google8BBF16 (native)maker’s file15.8 / 90.0 GB3–4 tok/s125KSystem memory
Runs, slowlyQwen2.5-7B-Instruct-1M Alibaba Qwen7.6BBF16 (native)maker’s file15.3 / 90.0 GB3–4 tok/s977KSystem memory
Runs, slowlyQwen2.5-7B-Instruct Alibaba Qwen7.6BBF16 (native)maker’s file15.3 / 90.0 GB3–4 tok/s32KSystem memory
Runs, slowlyDeepSeek-R1-Distill-Qwen-7B DeepSeek7.6BBF16 (native)maker’s file15.3 / 90.0 GB3–4 tok/s32KSystem memory
Runs, slowlyMistral-7B-Instruct-v0.1 Mistral AI7.2BBF16 (native)maker’s file15.2 / 90.0 GB3–4 tok/s32KSystem memory
Runs, slowlyWeVisOne-VL-4B Tencent4.8BBF16 (native)maker’s file10.6 / 90.0 GB4–6 tok/s256KSystem memory
Runs, slowlyQianfan-OCR Baidu4.7BBF16 (native)maker’s file10.5 / 90.0 GB4–6 tok/s31KSystem memory

Estimates from the makers’ Hugging Face repositories (as of 2026-11-06); filtered views aren’t indexed. How it’s estimated