AI Index

Open-weight models that run on Apple M6 (16GB) · 16 GB

  • Unified memory 16 GB
  • Bandwidth 153 GB/s
  • Usable 10.7 GB

Specs: Memory · Mac mini - Technical Specifications - Apple · Bandwidth · Mac mini - Technical Specifications - Apple

FitModelParameters / activePrecisionMemoryEst. speedLongest contextRuns as
Runs wellGLM-4.6V-Flash Zhipu AI10.3BIQ4_XS5.9 / 10.7 GB13–20 tok/s125KUnified memory
Runs wellQwen3.5-9B Alibaba Qwen9.7BQ4_05.8 / 10.7 GB13–20 tok/s163KUnified memory
Runs wellQianfan-VL-8B Baidu8.8BQ4_06.1 / 10.7 GB13–19 tok/s32KUnified memory
Runs wellGranite Switch 4.1 8B IBM8.8BIQ4_XS6.2 / 10.7 GB13–19 tok/s35KUnified memory
Runs wellQwen3-VL-8B-Instruct Alibaba Qwen8.8BIQ4_XS6.0 / 10.7 GB13–20 tok/s41KUnified memory
Runs wellQwen3-VL-8B-Thinking Alibaba Qwen8.8BIQ4_XS6.0 / 10.7 GB13–20 tok/s41KUnified memory
Runs wellpplx-embed-v2-context-9b Perplexity8.4BQ5_06.1 / 10.7 GB13–19 tok/s153KUnified memory
Runs wellQwen2.5-VL-7B-Instruct Alibaba Qwen8.3BQ4_K_M5.6 / 10.7 GB14–21 tok/s100KUnified memory
Runs wellQwen3-8B Alibaba Qwen8.2BMLX 4bitmaker’s file5.7 / 10.7 GB14–21 tok/s128KUnified memory
Runs wellDeepSeek-R1-Distill-Llama-8B DeepSeek8BQ4_K_M6.0 / 10.7 GB13–19 tok/s45KUnified memory
Runs wellGemma 4 E4B Google8BQ5_K_M5.9 / 10.7 GB13–20 tok/s125KUnified memory
Runs wellLing-3.0-tiny Ant Group7.9B / 1.3BQ8_0maker’s file8.5 / 10.7 GB47–71 tok/s88KUnified memory
Runs wellQwen2.5-7B-Instruct-1M Alibaba Qwen7.6BQ5_K_M6.0 / 10.7 GB13–19 tok/s93KUnified memory
Runs wellQwen2.5-7B-Instruct Alibaba Qwen7.6BQ5_K_Mmaker’s file6.0 / 10.7 GB13–19 tok/s32KUnified memory
Runs wellDeepSeek-R1-Distill-Qwen-7B DeepSeek7.6BQ5_K_M6.0 / 10.7 GB13–19 tok/s32KUnified memory
Runs wellMistral-7B-Instruct-v0.1 Mistral AI7.2BQ5_06.1 / 10.7 GB13–19 tok/s32KUnified memory
Runs wellGemma 4 E2B Google5.1BQ8_05.6 / 10.7 GB14–21 tok/s125KUnified memory
Runs wellWeVisOne-VL-4B Tencent4.8BFP86.1 / 10.7 GB13–19 tok/s40KUnified memory
Runs wellQianfan-OCR Baidu4.7BFP86.0 / 10.7 GB13–19 tok/s31KUnified memory
Runs wellQwen3.5-4B Alibaba Qwen4.7BQ8_05.4 / 10.7 GB15–22 tok/s177KUnified memory
Runs wellQwen3-VL-4B-Instruct Alibaba Qwen4.4BQ8_0maker’s file5.6 / 10.7 GB14–21 tok/s44KUnified memory
Runs wellQwen3-4B-Instruct-2507 Alibaba Qwen4BQ8_05.6 / 10.7 GB14–21 tok/s44KUnified memory
Runs wellQwen3-4B Alibaba Qwen4BMLX 8bitmaker’s file5.5 / 10.7 GB14–21 tok/s128KUnified memory
Runs wellDeepSeek-OCR DeepSeek3.3BBF16 (native)maker’s file7.2 / 10.7 GB27–41 tok/s8KUnified memory
Runs wellGranite-4.0-H-Micro IBM3.2BQ8_0maker’s file3.7 / 10.7 GB22–33 tok/s128KUnified memory
Runs wellQwen3-1.7B Alibaba Qwen2BBF16 (native)maker’s file5.2 / 10.7 GB15–23 tok/s40KUnified memory
Runs wellDeepSeek-R1-Distill-Qwen-1.5B DeepSeek1.8BBF16 (native)maker’s file4.0 / 10.7 GB20–30 tok/s125KUnified memory
Runs wellGLM-OCR Zhipu AI1.3BBF16 (native)maker’s file3.5 / 10.7 GB24–36 tok/s123KUnified memory
Runs wellQwen3-0.6B Alibaba Qwen0.8BBF16 (native)maker’s file2.8 / 10.7 GB31–47 tok/s40KUnified memory
Runs wellEmbeddingGemma 2 Google0.7BBF16 (native)maker’s file2.0 / 10.7 GB49–73 tok/s256KUnified memory
Runs wellQwen3-Embedding-0.6B Alibaba Qwen0.6BBF16 (native)maker’s file2.5 / 10.7 GB36–54 tok/s32KUnified memory
Runs wellbge-reranker-v2-m3 BAAI0.6BFP32 (native)maker’s file3.4 / 10.7 GB25–37 tok/s8KUnified memory
Runs wellbge-m3-unsupervised BAAI0.6BFP32 (native)maker’s file3.4 / 10.7 GB25–37 tok/s8KUnified memory
Runs wellbge-large-en-v1.5 BAAI0.3BFP32 (native)maker’s file1.8 / 10.7 GB55–82 tok/s512Unified memory
Runs wellbge-reranker-base BAAI0.3BFP32 (native)maker’s file1.6 / 10.7 GB68–101 tok/s514Unified memory
Runs wellbge-base-en-v1.5 BAAI0.1BFP32 (native)maker’s file0.9 / 10.7 GB167–251 tok/s512Unified memory
Runs wellbge-small-en-v1.5 BAAI0BFP32 (native)maker’s file0.6 / 10.7 GB535–803 tok/s512Unified memory
Runs, slowlyQwen2.5-14B-Instruct-1M Alibaba Qwen14.8BIQ4_XS9.3 / 10.7 GB8–12 tok/s15KUnified memory
Runs, slowlyQwen2.5-14B-Instruct Alibaba Qwen14.8BIQ4_XS9.3 / 10.7 GB8–12 tok/s15KUnified memory
Runs, slowlyDeepSeek-R1-Distill-Qwen-14B DeepSeek14.8BIQ4_XS9.3 / 10.7 GB8–12 tok/s15KUnified memory
Runs, slowlyQwen3-14B Alibaba Qwen14.8BMLX 4bitmaker’s file9.1 / 10.7 GB8–12 tok/s18KUnified memory
Runs, slowlyGemma 3 12B IT Google12.2BIQ4_XS9.3 / 10.7 GB8–12 tok/s11KUnified memory
Runs, slowlyQwen2.5-Omni-7B Alibaba Qwen10.7BQ6_K9.1 / 10.7 GB8–12 tok/s32KUnified memory

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