| Model | Verdict | Decode tok/s | Max ctx | Size | Chat | Coding | |
|---|---|---|---|---|---|---|---|
| Qwen/Qwen3-30B-A3B-Instruct-2507 lmstudio-community/Qwen3-30B-A3B-Instruct-2507-MLX-8bit · MLX-8BIT | Fits | 52.721.1–57.9 | 256K | 30.5B | p85 | – | |
| Qwen/Qwen3.5-122B-A10B unsloth/Qwen3.5-122B-A10B-GGUF · Q5_K_S | Fits | 50.020.0–55.0 | 256K | 125.1B | p77 | p49 | |
| Qwen/Qwen3.8-Flash-Next ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF · IQ3_S | Fits | 64.125.7–70.6 | 256K | 180B | p68 | p51 | |
| google/gemma-4-31B-it lmstudio-community/gemma-4-31B-it-MLX-8bit · MLX-8BIT | Fits | 12.14.90–13.4 | 256K | 31.3B | p63 | – | |
| MiniMaxAI/MiniMax-M2.5 unsloth/MiniMax-M2.5-GGUF · Q2_K_L | Fits | 39.415.7–43.3 | 131K | 228.7B | p46 | p61 | |
| Qwen/Qwen3.5-27B unsloth/Qwen3.5-27B-GGUF · Q8_0 | Fits | 15.26.10–16.7 | 256K | 27.8B | p60 | p46 | |
| Qwen/Qwen3.8-27B lmstudio-community/Qwen3.8-27B-MLX-8bit · MLX-8BIT | Fits | 15.06.00–16.5 | 256K | 27.8B | p59 | p33 | |
| Qwen/Qwen3.6-27B lmstudio-community/Qwen3.6-27B-MLX-8bit · MLX-8BIT | Fits | 15.06.00–16.5 | 256K | 27.8B | p51 | p54 | |
| nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16 unsloth/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF · MXFP4_MOE | Fits | 65.726.3–72.2 | 256K | 123.6B | p50 | p15 | |
| Qwen/Qwen3.5-35B-A3B unsloth/Qwen3.5-35B-A3B-GGUF · Q8_0 | Fits | 80.432.2–88.4 | 256K | 36B | p49 | p35 | |
| google/gemma-4-26B-A4B-it lmstudio-community/gemma-4-26B-A4B-it-MLX-8bit · MLX-8BIT | Fits | 66.626.7–73.3 | 256K | 25.8B | p49 | – | |
| openai/gpt-oss-120b unsloth/gpt-oss-120b-GGUF · F16 | Fits | 77.931.2–85.7 | 128K | 117B | p47 | p15 | |
| Qwen/Qwen3.5-9B lmstudio-community/Qwen3.5-9B-MLX-8bit · MLX-8BIT | Fits | 39.115.7–43.0 | 256K | 9.7B | p46 | – | |
| ornith-ai/Ornith-1.5-9B BF16 | Fits | 23.99.60–26.3 | 256K | 9.7B | p46 | p24 | |
| Qwen/Qwen3.6-35B-A3B lmstudio-community/Qwen3.6-35B-A3B-MLX-8bit · MLX-8BIT | Fits | 79.731.9–87.6 | 256K | 36B | p43 | p41 | |
| openbmb/MiniCPM-V-4.7-35B-A3B openbmb/MiniCPM-V-4.7-35B-A3B-gguf · Q8_0 | Fits | 80.632.2–88.6 | 256K | 35.2B | p42 | – | |
| ornith-ai/Ornith-1.0-35B lmstudio-community/Ornith-1.0-35B-MLX-8bit · MLX-8BIT | Fits | 79.831.9–87.8 | 256K | 35.1B | – | p40 | |
| openai/gpt-oss-20b MXFP4 | Fits | 62.324.9–68.5 | 128K | 21.1B | p39 | p15 | |
| Qwen/Qwen3.5-4B BF16 | Fits | 39.215.7–43.1 | 256K | 4.7B | p39 | – | |
| google/gemma-4-12B-it unsloth/gemma-4-12b-it-GGUF · Q8_0 | Fits | 26.222.3–30.1 | 256K | 12B | p38 | – | |
| Qwen/Qwen3-4B-Instruct-2507 BF16 | Fits | 36.030.6–41.4 | 256K | 4B | p38 | – | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 BF16 | Fits | 81.832.7–90.0 | 256K | 32.9B | p37 | p9 | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 MIXED_PRECISION | Fits | 96.638.7–106 | 1M | 47.1B | p36 | p9 | |
| poolside/Laguna-S-2.1 TensorFold/Laguna-S-2.1-MLX-4bit · MLX-4BIT | Fits | 64.025.6–70.4 | 697K | 117.6B | – | p34 | |
| nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 BF16 | Fits | 83.633.4–91.9 | 256K | 31.6B | p33 | – | |
| poolside/Laguna-XS-2.1 AtomicChat/Laguna-XS-2.1-MLX-8bit · MLX-8BIT | Fits | 87.435.0–96.2 | 256K | 33.4B | – | p30 | |
| Qwen/Qwen2.5-7B BF16 | Fits | 26.422.4–30.4 | 128K | 7.6B | p29 | – | |
| google/gemma-4-E4B-it lmstudio-community/gemma-4-E4B-it-MLX-8bit · MLX-8BIT | Fits | 52.821.1–58.1 | 128K | 8B | p27 | – | |
| openbmb/MiniCPM5-2B BF16 | Fits | 51.744.0–59.5 | 128K | 2.5B | p27 | p3 | |
| poolside/Laguna-XS.2 BF16 | Fits | 54.421.8–59.8 | 256K | 33.4B | – | p25 | |
| Qwen/Qwen3.5-2B BF16 | Fits | 67.326.9–74.0 | 256K | 2.3B | p24 | – | |
| ornith-ai/Ornith-1.0-9B lmstudio-community/Ornith-1.0-9B-MLX-8bit · MLX-8BIT | Fits | 39.215.7–43.1 | 256K | 9.4B | – | p20 | |
| google/gemma-4-E2B-it lmstudio-community/gemma-4-E2B-it-MLX-8bit · MLX-8BIT | Fits | 68.927.6–75.8 | 128K | 5.1B | p18 | – | |
| Qwen/Qwen2.5-1.5B BF16 | Fits | 72.061.2–82.8 | 128K | 1.5B | p15 | – | |
| mistralai/Mistral-7B-Instruct-v0.2 BF16 | Fits | 25.221.4–29.0 | 32K | 7.2B | p14 | – | |
| Qwen/Qwen2.5-0.5B BF16 | Fits | 125106–144 | 32K | 0.5B | p5 | – | |
| Qwen/Qwen3.5-0.8B BF16 | Fits | 11043.8–121 | 256K | 0.9B | p4 | – | |
| incoai/Qwen3.8-27B-DFlash2 BF16 | Fits | 136116–157 | 256K | 1.9B | – | – | |
| Qwen/Qwen2.5-0.5B-Instruct BF16 | Fits | 125106–144 | 32K | 0.5B | – | – | |
| datalab-to/surya-ocr-2 BF16 | Fits | 12047.9–132 | 256K | 0.7B | – | – | |
| Qwen/Qwen3.5-0.8B-Base BF16 | Fits | 11043.8–121 | 256K | 0.9B | – | – | |
| baidu/Unlimited-OCR-2 AtomicChat/Unlimited-OCR-2-GGUF · BF16 | Fits | 10943.5–120 | 128K | 3.3B | – | – | |
| nvidia/Qwen3.6-35B-A3B-NVFP4 MIXED_PRECISION | Fits | 95.438.2–105 | 256K | 34B | – | – | |
| ornith-ai/Ornith-1.5-35B-A3B-NVFP4 MIXED_PRECISION | Fits | 95.438.2–105 | 256K | 34B | – | – | |
| openbmb/MiniCPM-V-4.7-1B BF16 | Fits | 91.936.8–101 | 256K | 1.6B | – | – | |
| Qwen/Qwen2.5-1.5B-Instruct Qwen/Qwen2.5-1.5B-Instruct-GGUF · Q8_0 | Fits | 87.374.2–100 | 32K | 1.5B | – | – | |
| Qwen/Qwen3-0.6B-Base BF16 | Fits | 86.873.7–99.8 | 32K | 0.6B | – | – | |
| Qwen/Qwen3-0.6B BF16 | Fits | 82.269.9–94.5 | 40K | 0.8B | – | – | |
| nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16 unsloth/NVIDIA-Nemotron-3-Nano-4B-GGUF · Q8_0 | Fits | 76.230.5–83.8 | 256K | 4B | – | – | |
| ornith-ai/Ornith-1.0-35B-GGUF Q8_0 | Fits | 72.629.1–79.9 | 256K | 34.7B | – | – | |
| ornith-ai/Ornith-1.5-35B-A3B-GGUF Q8_0 | Fits | 71.928.8–79.1 | 256K | 35.5B | – | – | |
| baidu/Unlimited-OCR BF16 | Fits | 62.124.8–68.3 | 32K | 3.3B | – | – | |
| Qwen/Qwen3-1.7B-Base BF16 | Fits | 62.052.7–71.3 | 32K | 1.7B | – | – | |
| Qwen/Qwen3-VL-2B-Instruct Qwen/Qwen3-VL-2B-Instruct-GGUF · F16 | Fits | 61.952.6–71.2 | 256K | 2.1B | – | – | |
| Qwen/Qwen2-VL-2B-Instruct BF16 | Fits | 60.251.1–69.2 | 32K | 2.2B | – | – | |
| ornith-ai/Ornith-1.5-35B-A3B-FP8 COMPRESSED-TENSORS | Fits | 59.923.9–65.8 | 256K | 36B | – | – | |
| Qwen/Qwen2.5-3B-Instruct Qwen/Qwen2.5-3B-Instruct-GGUF · Q8_0 | Fits | 59.850.8–68.8 | 32K | 3.1B | – | – | |
| Qwen/Qwen3-1.7B BF16 | Fits | 57.448.8–66.0 | 40K | 2B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct-FP8 FP8 | Fits | 52.821.1–58.1 | 256K | 30.5B | – | – | |
| prism-ml/Bonsai-27B-mlx-1bit MLX-1BIT | Fits | 52.320.9–57.6 | 256K | 26.4B | – | – | |
| openbmb/MiniCPM5-2B-GGUF F16 | Fits | 51.743.9–59.5 | 128K | 2.5B | – | – | |
| Qwen/Qwen3-VL-4B-Instruct lmstudio-community/Qwen3-VL-4B-Instruct-MLX-8bit · MLX-8BIT | Fits | 46.018.4–50.6 | 256K | 4.4B | – | – | |
| Qwen/Qwen2.5-VL-3B-Instruct unsloth/Qwen2.5-VL-3B-Instruct-GGUF · BF16 | Fits | 45.538.7–52.3 | 125K | 3.8B | – | – | |
| ornith-ai/Ornith-1.5-9B-NVFP4 MIXED_PRECISION | Fits | 44.517.8–49.0 | 256K | 9B | – | – | |
| ornith-ai/Ornith-1.0-9B-GGUF Q8_0 | Fits | 41.616.7–45.8 | 256K | 9B | – | – | |
| Qwen/Qwen2.5-7B-Instruct Qwen/Qwen2.5-7B-Instruct-GGUF · Q8_0 | Fits | 41.235.0–47.3 | 32K | 7.6B | – | – | |
| ornith-ai/Ornith-1.5-122B-GGUF Q5_K_M | Fits | 41.116.4–45.2 | 256K | 124.6B | – | – | |
| Qwen/Qwen2.5-VL-7B-Instruct unsloth/Qwen2.5-VL-7B-Instruct-GGUF · Q8_0 | Fits | 41.034.9–47.2 | 125K | 8.3B | – | – | |
| Qwen/Qwen2-VL-7B-Instruct ggml-org/Qwen2-VL-7B-Instruct-GGUF · Q8_0 | Fits | 41.034.9–47.2 | 32K | 8.3B | – | – | |
| ornith-ai/Ornith-1.5-9B-GGUF Q8_0 | Fits | 40.716.3–44.8 | 256K | 9.2B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct BF16 | Fits | 40.016.0–43.9 | 256K | 30.5B | – | – | |
| Qwen/Qwen3-30B-A3B BF16 | Fits | 40.016.0–43.9 | 40K | 30.5B | – | – | |
| Qwen/Qwen3.5-4B-Base BF16 | Fits | 39.215.7–43.1 | 256K | 4.7B | – | – | |
| prism-ml/Ternary-Bonsai-27B-mlx-2bit MLX-2BIT | Fits | 39.015.6–42.8 | 256K | 26.4B | – | – | |
| HauhauCS/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive Q8_K_P | Fits | 37.915.2–41.7 | 128K | 7.5B | – | – | |
| Qwen/Qwen3-8B-Base Qwen/Qwen3-Embedding-8B-GGUF · Q8_0 | Fits | 37.732.1–43.4 | 32K | 8.2B | – | – | |
| Qwen/Qwen2.5-Coder-7B-Instruct Qwen/Qwen2.5-Coder-7B-Instruct-GGUF · Q4_K_M | Fits | 37.431.8–43.1 | 32K | 7.6B | – | – | |
| Qwen/Qwen3-8B lmstudio-community/Qwen3-8B-MLX-8bit · MLX-8BIT | Fits | 36.014.4–39.6 | 40K | 8.2B | – | – | |
| Qwen/Qwen3-4B BF16 | Fits | 36.030.6–41.4 | 40K | 4B | – | – | |
| datalab-to/chandra-ocr-2 BF16 | Fits | 35.514.2–39.1 | 256K | 5.3B | – | – | |
| Qwen/Qwen3-VL-8B-Instruct lmstudio-community/Qwen3-VL-8B-Instruct-MLX-8bit · MLX-8BIT | Fits | 33.313.3–36.6 | 256K | 8.8B | – | – | |
| Qwen/Qwen3-14B lmstudio-community/Qwen3-14B-MLX-8bit · MLX-8BIT | Fits | 23.39.30–25.6 | 40K | 14.8B | – | – | |
| Qwen/Qwen2.5-14B-Instruct Qwen/Qwen2.5-14B-Instruct-GGUF · Q8_0 | Fits | 22.419.1–25.8 | 32K | 14.8B | – | – | |
| Qwen/Qwen2.5-Coder-14B-Instruct Qwen/Qwen2.5-Coder-14B-Instruct-GGUF · Q4_K_M | Fits | 20.417.3–23.5 | 32K | 14.8B | – | – | |
| RadixArk/Qwen3.8-27B-NVFP4 MIXED_PRECISION | Fits | 19.37.70–21.2 | 256K | 26.4B | – | – | |
| Qwen/Qwen2.5-32B-Instruct lmstudio-community/Qwen2.5-32B-Instruct-MLX-8bit · MLX-8BIT | Fits | 11.54.60–12.6 | 32K | 32.8B | – | – | |
| Qwen/Qwen2.5-Coder-32B-Instruct lmstudio-community/Qwen2.5-Coder-32B-Instruct-MLX-8bit · MLX-8BIT | Fits | 11.54.60–12.6 | 32K | 32.8B | – | – | |
| farbodtavakkoli/OTel-2.0-LLM-31B-IT wfakhri/OTel-2.0-LLM-31B-IT-GGUF · Q8_0 | Fits | 11.39.60–13.0 | 256K | 31.3B | – | – | |
| Qwen/Qwen3-32B lmstudio-community/Qwen3-32B-MLX-8bit · MLX-8BIT | Fits | 11.24.50–12.3 | 40K | 32.8B | – | – | |
| nvidia/Gemma-4-31B-IT-NVFP4 NVFP4 | Fits | – | 256K | 29.2B | – | – | |
| nvidia/Gemma-4-26B-A4B-NVFP4 NVFP4 | Fits | – | 256K | 24.5B | – | – | |
| nvidia/Qwen2.5-VL-7B-Instruct-NVFP4 NVFP4 | Fits | – | 125K | 8.2B | – | – |
Quality = benchmark percentile against all indexed models (p50 = median). Speeds are single-user decode ranges.