Apple M1 Pro

MacBook Pro

16 GB RAM · 63 models tested

Avg Score
70 /100
Avg Speed
39 tok/s
Models Tested
63
Model Size Score tok/s TTFT Verdict
nemotron-3-nano:4b 4.0B Q4_K_M 93 31.7 2.9s Excellent
gemma4:e4b-it-q8_0 8.0B Q8_0 91 30.0 6.2s Excellent
gemma4:e4b-mlx 90 41.8 4.2s Excellent
gemma4:e4b 8.0B Q4_K_M 90 35.6 6.1s Excellent
gemma4:e4b 8.0B Q4_K_M 90 35.5 6.1s Excellent
gemma4:e4b 8.0B Q4_K_M 90 35.7 6.5s Excellent
hf.co/unsloth/gemma-4-E4B-it-GGUF:Q5_K_M 7.52B unknown 88 28.3 224 ms Excellent
maternion/ling-3.0-tiny:8b 7.9B Q4_K_M 88 73.0 175 ms Excellent
gemma4:e4b-it-q8_0 8.0B Q8_0 88 27.8 204 ms Excellent
hf.co/lmstudio-community/gemma-4-E4B-it-GGUF:Q6_K 7.52B unknown 88 34.6 210 ms Excellent
gemma4:e2b-mlx 88 72.8 30.0s Not Rec.
hf.co/ornith-ai/Ornith-1.0-9B-GGUF:Q4_K_M 8.95B unknown 87 22.2 334 ms Excellent
hf.co/unsloth/gemma-4-E4B-it-GGUF:Q4_K_M 7.52B unknown 87 35.0 204 ms Excellent
hf.co/unsloth/gpt-oss-20b-GGUF:Q4_K_M 20.9B unknown 86 35.9 5.2s Excellent
nemotron-3-nano:4b 4.0B Q4_K_M 85 31.7 265 ms Excellent
qwen3.5:9b-mlx nvfp4 84 25.0 357 ms Excellent
hf.co/unsloth/gemma-4-E2B-it-GGUF:UD-Q8_K_XL 4.65B unknown 84 48.0 167 ms Excellent
qwen3.5:4b 4.7B Q4_K_M 83 36.6 204 ms Excellent
gemma4:e2b-mlx 82 70.4 96 ms Excellent
granite4.2:3b 3.7B Q4_K_M 80 46.8 135 ms Excellent
rnj-1:8b 8.3B Q4_K_M 80 22.3 404 ms Excellent
lfm2.5:8b 8.5B Q4_K_M 80 98.5 2.5s Excellent
hf.co/empero-ai/Qwen3.8-2B-Distill-GGUF:BF16 1.94B BF16 76 40.3 101 ms Good
hf.co/bartowski/ai9stars_G9v3-3B-GGUF:BF16 2.99B BF16 76 29.9 105 ms Good
ministral-3:8b 8.9B Q4_K_M 72 19.8 269 ms Good
llama3.1:8b 8.0B Q4_K_M 72 21.5 368 ms Good
gemma4:12b-mlx 71 25.0 30.0s Not Rec.
hf.co/openbmb/MiniCPM5-1B-GGUF:F16 1.08B F16 71 88.1 38 ms Good
ornith-1.5:9b 9.0B Q4_K_M 69 22.2 341 ms Good
qwen3.5:4b 4.7B Q4_K_M 69 36.7 6.6s Good
qwen3.5:4b 4.7B Q4_K_M 69 36.9 6.9s Good
hf.co/bloomer010/Ling-3.0-tiny-GGUF:Q8_0 7.89B unknown 68 58.4 220 ms Good
maternion/ling-3.0-tiny:8b 7.9B Q4_K_M 68 73.6 190 ms Good
hf.co/bloomer010/Ling-3.0-tiny-GGUF:Q8_0 7.89B unknown 68 58.9 208 ms Good
qwen3.5:4b-mlx nvfp4 68 42.2 4.4s Good
qwen3.5:4b 4.7B Q4_K_M 68 36.5 6.6s Good
maternion/ling-3.0-tiny:8b 7.9B Q4_K_M 68 72.1 175 ms Good
hf.co/empero-ai/Qwen3.8-2B-Distill-GGUF:BF16 1.94B BF16 66 40.4 93 ms Good
hf.co/empero-ai/Qwen3.8-2B-Distill-GGUF:BF16 1.94B BF16 66 40.2 325 ms Good
hf.co/empero-ai/Qwen3.8-2B-Distill-GGUF:BF16 1.94B BF16 66 40.2 101 ms Good
hf.co/bartowski/ai9stars_G9v3-3B-GGUF:BF16 2.99B BF16 65 29.6 264 ms Good
hf.co/bartowski/ai9stars_G9v3-3B-GGUF:BF16 2.99B BF16 64 29.9 100 ms Good
hf.co/bartowski/ai9stars_G9v3-3B-GGUF:BF16 2.99B BF16 64 29.9 108 ms Good
gemma4:12b 11.9B Q4_K_M 63 15.3 30.0s Not Rec.
hf.co/openbmb/MiniCPM5-1B-GGUF:F16 1.08B F16 62 88.6 165 ms Good
hf.co/ornith-ai/Ornith-1.0-9B-GGUF:Q4_K_M 8.95B unknown 62 22.3 324 ms Good
hf.co/openbmb/MiniCPM5-1B-GGUF:F16 1.08B F16 62 89.1 42 ms Good
hf.co/openbmb/MiniCPM5-1B-GGUF:F16 1.08B F16 62 88.0 36 ms Good
hf.co/LiquidAI/LFM2.5-2.6B-GGUF:F16 2.7B unknown 61 31.4 138 ms Good
hf.co/ornith-ai/Ornith-1.0-9B-GGUF:Q4_K_M 8.95B unknown 61 22.2 352 ms Good
hf.co/LiquidAI/LFM2.5-2.6B-GGUF:F16 2.7B unknown 61 31.0 129 ms Good
hf.co/LiquidAI/LFM2.5-2.6B-GGUF:F16 2.7B unknown 61 31.3 131 ms Good
granite4.2:3b 3.7B Q4_K_M 59 46.3 167 ms Marginal
hf.co/InternScience/Agents-A1-4B-Q8_0-GGUF:Q8_0 4.21B unknown 59 30.9 891 ms Marginal
qwen3.5:9b 9.7B Q4_K_M 58 22.1 9.1s Marginal
hf.co/ornith-ai/Ornith-1.0-9B-GGUF:Q8_0 8.95B unknown 57 18.6 303 ms Marginal
qwen3.5:9b-mlx nvfp4 54 25.2 6.3s Marginal
hf.co/Edge-Quant/Nanbeige4.1-3B-Q8_0-GGUF:Q8_0 3.93B unknown 52 43.5 30.0s Not Rec.
hf.co/unsloth/Qwen3.8-27B-GGUF:UD-IQ1_S 27.3B unknown 46 6.0 2.2s Marginal
hf.co/mradermacher/gemma-4-E4B-GGUF:Q6_K 7.52B unknown 41 34.6 179 ms Marginal
hf.co/unsloth/Qwen3.8-27B-GGUF:UD-IQ2_XXS 27.3B unknown 31 5.4 5.1s Not Rec.
hf.co/unsloth/Qwen3.8-27B-GGUF:UD-IQ1_S 27.3B unknown 22 6.1 3.3s Not Rec.
hf.co/unsloth/Qwen3.8-27B-GGUF:UD-IQ1_S 27.3B unknown 22 6.1 3.2s Not Rec.