Open-weight models that run on Apple M5 Ultra (64-core GPU) · 256 GB
- Unified memory 256 GB
- Bandwidth 1200 GB/s
- Usable 192.0 GB
Specs: Memory · Mac Studio - Technical Specifications - Apple · Bandwidth · Mac Studio - Technical Specifications - Apple
| Fit | Model | Parameters / active | Precision | Memory | Est. speed | Longest context | Runs as |
|---|---|---|---|---|---|---|---|
| ●Runs well | GLM-4.5 Zhipu AI | 358.3B | IQ4_XS | 25–38 tok/s | 16K | Unified memory | |
| ●Runs well | GLM-4.7 Zhipu AI | 358.3B | IQ4_XS | 25–38 tok/s | 16K | Unified memory | |
| ●Runs well | GLM-4.6 Zhipu AI | 356.8B | IQ4_XS | 26–39 tok/s | 18K | Unified memory | |
| ●Runs well | GLM-5.3-Flash Zhipu AI | 321.3B / 18B | IQ4_XS | 60–91 tok/s | 570K | Unified memory | |
| ●Runs well | MiMo-V2.5 Xiaomi | 310.8B / 15B | Q4_0 | 62–93 tok/s | 150K | Unified memory | |
| ●Runs well | DeepSeek-V4-Flash (2026-07-31) DeepSeek | 304.2B | FP8 (native)maker’s file | 31–47 tok/s | 343K | Unified memory | |
| ●Runs well | Hy3 Tencent | 298.8B / 21B | Q4_0 | 41–62 tok/s | 88K | Unified memory | |
| ●Runs well | DeepSeek-V4-Flash (2026-04-24) DeepSeek | 290.9B / 13B | FP8 (native)maker’s file | 76–115 tok/s | 428K | Unified memory | |
| ●Runs well | DeepSeek-V4-Flash DeepSeek | 290.9B / 13B | FP8 (native)maker’s file | 76–115 tok/s | 427K | Unified memory | |
| ●Runs well | ERNIE-5.5-Flash-Base Baidu | 286.2B | Q4_K_M | speed not estimated | 465K | Unified memory | |
| ●Runs well | Qwen3-VL-235B-A22B-Instruct Alibaba Qwen | 235.7B | Q5_K_M | 34–51 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-VL-235B-A22B-Thinking Alibaba Qwen | 235.7B | Q5_K_M | 34–51 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-235B-A22B-Instruct-2507 Alibaba Qwen | 235.1B / 22B | Q5_K_M | 35–52 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3-235B-A22B-Thinking-2507 Alibaba Qwen | 235.1B / 22B | Q5_K_M | 35–52 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3-235B-A22B Alibaba Qwen | 235.1B / 22B | Q5_K_Mmaker’s file | 35–52 tok/s | 128K | Unified memory | |
| ●Runs well | MiniMax-M2.1 MiniMax | 228.7B | Q5_K_M | 60–91 tok/s | 136K | Unified memory | |
| ●Runs well | MiniMax-M2.5 MiniMax | 228.7B | Q5_K_M | 60–91 tok/s | 136K | Unified memory | |
| ●Runs well | MiniMax-M2.7 MiniMax | 228.7B | Q5_K_M | 60–91 tok/s | 136K | Unified memory | |
| ●Runs well | MiniMax-M2 MiniMax | 228.7B / 10B | Q5_K_M | 65–98 tok/s | 136K | Unified memory | |
| ●Runs well | MiniMax-M3.1-Flash MiniMax | 220.5B / 23B | Q5_K_M | 35–52 tok/s | 330K | Unified memory | |
| ●Runs well | Step 3.7 Flash StepFun | 201.4B | Q6_K | 145–218 tok/s | 172K | Unified memory | |
| ●Runs well | Step 3.5 Flash StepFun | 199.4B / 11B | Q6_K | 57–85 tok/s | 181K | Unified memory | |
| ●Runs well | Ming-3.0-audio-flash Ant Group | 128.2B / 6.5B | Q8_0 | 82–123 tok/s | 128K | Unified memory | |
| ●Runs well | Ling-3.0-flash Ant Group | 127.5B / 5.1B | Q8_0maker’s file | 103–155 tok/s | 250K | Unified memory | |
| ●Runs well | Qwen3.5-122B-A10B Alibaba Qwen | 125.1B / 10B | Q8_0 | 55–83 tok/s | 256K | Unified memory | |
| ●Runs well | Ling-3.0-flash-VL Ant Group | 124.8B / 5.5B | Q8_0 | 96–144 tok/s | 128K | Unified memory | |
| ●Runs well | gpt-oss-120b OpenAI | 116.8B / 5.1B | MXFP4 (native)maker’s file | 190–285 tok/s | 128K | Unified memory | |
| ●Runs well | GLM-4.5-Air Zhipu AI | 110.5B | Q8_0 | 31–46 tok/s | 125K | Unified memory | |
| ●Runs well | GLM-4.5V Zhipu AI | 107.7B / 12B | Q8_0 | 42–63 tok/s | 63K | Unified memory | |
| ●Runs well | GLM-4.6V Zhipu AI | 107.7B | Q8_0 | 36–54 tok/s | 125K | Unified memory | |
| ●Runs well | Ling-2.6-flash Ant Group | 107.5B / 7.4B | Q8_0 | 73–110 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-Next-80B-A3B-Instruct Alibaba Qwen | 81.3B / 3B | BF16 (native)maker’s file | 97–145 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3-Next-80B-A3B-Thinking Alibaba Qwen | 81.3B / 3B | BF16 (native)maker’s file | 97–145 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3-Coder-Next Alibaba Qwen | 79.7B / 3B | BF16 (native)maker’s file | 97–145 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen2.5-VL-72B-Instruct Alibaba Qwen | 73.4B | Q4_K_M | 13–19 tok/s | 125K | Unified memory | |
| ●Runs well | Qwen2-VL-72B-Instruct Alibaba Qwen | 73.4B | Q4_K_M | 13–19 tok/s | 32K | Unified memory | |
| ●Runs well | Qwen2.5-72B-Instruct Alibaba Qwen | 72.7B | Q4_K_Mmaker’s file | 13–19 tok/s | 32K | Unified memory | |
| ●Runs well | Qianfan-VL-70B Baidu | 71.8B | Q4_K_M | 13–19 tok/s | 32K | Unified memory | |
| ●Runs well | DeepSeek-R1-Distill-Llama-70B DeepSeek | 70.6B | Q4_K_M | 13–20 tok/s | 125K | Unified memory | |
| ●Runs well | Qwen3.5-35B-A3B Alibaba Qwen | 36B / 3B | BF16 (native)maker’s file | 97–146 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3.6-35B-A3B Alibaba Qwen | 36B / 3B | BF16 (native)maker’s file | 97–146 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen2.5-VL-32B-Instruct Alibaba Qwen | 33.5B | Q8_0 | 16–24 tok/s | 125K | Unified memory | |
| ●Runs well | Qwen3-VL-32B-Instruct Alibaba Qwen | 33.4B | Q8_0maker’s file | 16–24 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-VL-32B-Thinking Alibaba Qwen | 33.4B | Q8_0maker’s file | 16–24 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen2.5-32B-Instruct Alibaba Qwen | 32.8B | Q8_0maker’s file | 16–24 tok/s | 32K | Unified memory | |
| ●Runs well | Qwen2.5-Coder-32B-Instruct Alibaba Qwen | 32.8B | INT8 (GPTQ)maker’s file | 16–24 tok/s | 32K | Unified memory | |
| ●Runs well | QwQ-32B Alibaba Qwen | 32.8B | Q8_0maker’s file | 16–24 tok/s | 40K | Unified memory | |
| ●Runs well | DeepSeek-R1-Distill-Qwen-32B DeepSeek | 32.8B | Q8_0 | 16–24 tok/s | 32K | Unified memory | |
| ●Runs well | Qwen3-32B Alibaba Qwen | 32.8B | MLX 8bitmaker’s file | 17–25 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-Omni-30B-A3B-Captioner Alibaba Qwen | 31.7B | BF16 (native)maker’s file | 61–91 tok/s | 64K | Unified memory | |
| ●Runs well | Gemma 4 31B IT Google | 31.3B | Q8_0 | 17–25 tok/s | 250K | Unified memory | |
| ●Runs well | GLM-4.7-Flash Zhipu AI | 31.2B | BF16 (native)maker’s file | 56–83 tok/s | 195K | Unified memory | |
| ●Runs well | Qwen3-VL-30B-A3B-Instruct Alibaba Qwen | 31.1B | BF16 (native)maker’s file | 70–105 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-VL-30B-A3B-Thinking Alibaba Qwen | 31.1B | BF16 (native)maker’s file | 70–105 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-30B-A3B-Instruct-2507 Alibaba Qwen | 30.5B / 3.3B | BF16 (native)maker’s file | 81–122 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3-30B-A3B-Thinking-2507 Alibaba Qwen | 30.5B / 3.3B | BF16 (native)maker’s file | 81–122 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3-30B-A3B Alibaba Qwen | 30.5B / 3.3B | BF16 (native)maker’s file | 81–122 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-Coder-30B-A3B-Instruct Alibaba Qwen | 30.5B / 3.3B | BF16 (native)maker’s file | 81–122 tok/s | 256K | Unified memory | |
| ●Runs well | Hy-Memory-30B-A3B Tencent | 30.1B / 3B | BF16 (native)maker’s file | 88–132 tok/s | 63K | Unified memory | |
| ●Runs well | Granite Switch 4.1 30B IBM | 30B | Q8_0maker’s file | 17–25 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3.5-27B Alibaba Qwen | 27.8B | Q8_0 | 20–30 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3.6-27B Alibaba Qwen | 27.8B | Q8_0 | 20–30 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3.8-27B Alibaba Qwen | 27.8B | Q8_0 | 20–30 tok/s | 256K | Unified memory | |
| ●Runs well | Gemma 4 26B A4B IT Google | 25.8B / 3.8B | BF16 (native)maker’s file | 74–110 tok/s | 250K | Unified memory | |
| ●Runs well | Magistral Small 1.2 Mistral AI | 24B | Q8_0maker’s file | 23–34 tok/s | 125K | Unified memory | |
| ●Runs well | Devstral Small 1.0 Mistral AI | 23.6B | Q8_0 | 23–34 tok/s | 125K | Unified memory | |
| ●Runs well | Devstral Small 1.1 Mistral AI | 23.6B | Q8_0 | 23–34 tok/s | 125K | Unified memory | |
| ●Runs well | Magistral Small 1.0 Mistral AI | 23.6B | Q8_0 | 23–34 tok/s | 39K | Unified memory | |
| ●Runs well | Magistral Small 1.1 Mistral AI | 23.6B | Q8_0maker’s file | 23–34 tok/s | 39K | Unified memory | |
| ●Runs well | ERNIE-4.5-21B-A3B-Thinking Baidu | 21.8B / 3B | BF16 (native)maker’s file | 93–139 tok/s | 128K | Unified memory | |
| ●Runs well | GPT OSS 20B OpenAI | 20.9B / 3.6B | MXFP4 (native)maker’s file | 233–350 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen2.5-14B-Instruct-1M Alibaba Qwen | 14.8B | BF16 (native)maker’s file | 19–29 tok/s | 870K | Unified memory | |
| ●Runs well | Qwen2.5-14B-Instruct Alibaba Qwen | 14.8B | BF16 (native)maker’s file | 19–29 tok/s | 32K | Unified memory | |
| ●Runs well | DeepSeek-R1-Distill-Qwen-14B DeepSeek | 14.8B | BF16 (native)maker’s file | 19–29 tok/s | 32K | Unified memory | |
| ●Runs well | Qwen3-14B Alibaba Qwen | 14.8B | BF16 (native)maker’s file | 19–29 tok/s | 128K | Unified memory | |
| ●Runs well | Gemma 3 12B IT Google | 12.2B | BF16 (native)maker’s file | 22–33 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen2.5-Omni-7B Alibaba Qwen | 10.7B | BF16 (native)maker’s file | 26–39 tok/s | 32K | Unified memory | |
| ●Runs well | GLM-4.6V-Flash Zhipu AI | 10.3B | BF16 (native)maker’s file | 29–43 tok/s | 125K | Unified memory | |
| ●Runs well | Qwen3.5-9B Alibaba Qwen | 9.7B | BF16 (native)maker’s file | 31–46 tok/s | 256K | Unified memory | |
| ●Runs well | Qianfan-VL-8B Baidu | 8.8B | BF16 (native)maker’s file | 32–48 tok/s | 32K | Unified memory | |
| ●Runs well | Granite Switch 4.1 8B IBM | 8.8B | BF16 (native)maker’s file | 32–47 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-VL-8B-Instruct Alibaba Qwen | 8.8B | BF16 (native)maker’s file | 32–48 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-VL-8B-Thinking Alibaba Qwen | 8.8B | BF16 (native)maker’s file | 32–48 tok/s | 128K | Unified memory | |
| ●Runs well | pplx-embed-v2-context-9b Perplexity | 8.4B | FP32 (native)maker’s file | 18–27 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen2.5-VL-7B-Instruct Alibaba Qwen | 8.3B | BF16 (native)maker’s file | 35–53 tok/s | 125K | Unified memory | |
| ●Runs well | Qwen3-8B Alibaba Qwen | 8.2B | BF16 (native)maker’s file | 34–51 tok/s | 128K | Unified memory | |
| ●Runs well | DeepSeek-R1-Distill-Llama-8B DeepSeek | 8B | BF16 (native)maker’s file | 35–53 tok/s | 125K | Unified memory | |
| ●Runs well | Gemma 4 E4B Google | 8B | BF16 (native)maker’s file | 37–56 tok/s | 125K | Unified memory | |
| ●Runs well | Ling-3.0-tiny Ant Group | 7.9B / 1.3B | BF16 (native)maker’s file | 212–318 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen2.5-7B-Instruct-1M Alibaba Qwen | 7.6B | BF16 (native)maker’s file | 38–57 tok/s | 977K | Unified memory | |
| ●Runs well | Qwen2.5-7B-Instruct Alibaba Qwen | 7.6B | BF16 (native)maker’s file | 38–57 tok/s | 32K | Unified memory | |
| ●Runs well | DeepSeek-R1-Distill-Qwen-7B DeepSeek | 7.6B | BF16 (native)maker’s file | 38–57 tok/s | 32K | Unified memory | |
| ●Runs well | Mistral-7B-Instruct-v0.1 Mistral AI | 7.2B | BF16 (native)maker’s file | 39–58 tok/s | 32K | Unified memory | |
| ●Runs well | Gemma 4 E2B Google | 5.1B | BF16 (native)maker’s file | 58–87 tok/s | 125K | Unified memory | |
| ●Runs well | WeVisOne-VL-4B Tencent | 4.8B | BF16 (native)maker’s file | 55–83 tok/s | 256K | Unified memory | |
| ●Runs well | Qianfan-OCR Baidu | 4.7B | BF16 (native)maker’s file | 56–84 tok/s | 31K | Unified memory | |
| ●Runs well | Qwen3.5-4B Alibaba Qwen | 4.7B | BF16 (native)maker’s file | 63–94 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3-VL-4B-Instruct Alibaba Qwen | 4.4B | BF16 (native)maker’s file | 60–89 tok/s | 250K | Unified memory | |
| ●Runs well | Qwen3-4B-Instruct-2507 Alibaba Qwen | 4B | BF16 (native)maker’s file | 65–97 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3-4B Alibaba Qwen | 4B | BF16 (native)maker’s file | 65–97 tok/s | 128K | Unified memory | |
| ●Runs well | DeepSeek-OCR DeepSeek | 3.3B | BF16 (native)maker’s file | 215–323 tok/s | 8K | Unified memory | |
| ●Runs well | Granite-4.0-H-Micro IBM | 3.2B | BF16 (native)maker’s file | 93–140 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-1.7B Alibaba Qwen | 2B | BF16 (native)maker’s file | 120–180 tok/s | 40K | Unified memory | |
| ●Runs well | DeepSeek-R1-Distill-Qwen-1.5B DeepSeek | 1.8B | BF16 (native)maker’s file | 158–238 tok/s | 125K | Unified memory | |
| ●Runs well | GLM-OCR Zhipu AI | 1.3B | BF16 (native)maker’s file | 188–282 tok/s | 128K | Unified memory | |
| ●Runs well | Qwen3-0.6B Alibaba Qwen | 0.8B | BF16 (native)maker’s file | 246–368 tok/s | 40K | Unified memory | |
| ●Runs well | EmbeddingGemma 2 Google | 0.7B | BF16 (native)maker’s file | 380–571 tok/s | 256K | Unified memory | |
| ●Runs well | Qwen3-Embedding-0.6B Alibaba Qwen | 0.6B | BF16 (native)maker’s file | 282–422 tok/s | 32K | Unified memory | |
| ●Runs well | bge-reranker-v2-m3 BAAI | 0.6B | FP32 (native)maker’s file | 195–293 tok/s | 8K | Unified memory | |
| ●Runs well | bge-m3-unsupervised BAAI | 0.6B | FP32 (native)maker’s file | 195–293 tok/s | 8K | Unified memory | |
| ●Runs well | bge-large-en-v1.5 BAAI | 0.3B | FP32 (native)maker’s file | 431–647 tok/s | 512 | Unified memory | |
| ●Runs well | bge-reranker-base BAAI | 0.3B | FP32 (native)maker’s file | 530–796 tok/s | 514 | Unified memory | |
| ●Runs well | bge-base-en-v1.5 BAAI | 0.1B | FP32 (native)maker’s file | 1313–1970 tok/s | 512 | Unified memory | |
| ●Runs well | bge-small-en-v1.5 BAAI | 0B | FP32 (native)maker’s file | 4199–6298 tok/s | 512 | Unified memory |
Estimates from the makers’ Hugging Face repositories (as of 2026-11-06); filtered views aren’t indexed. How it’s estimated