| Model | Verdict | Decode tok/s | Max ctx | Size | Chat | Coding | |
|---|---|---|---|---|---|---|---|
| Qwen/Qwen3-30B-A3B-Instruct-2507 unsloth/Qwen3-30B-A3B-Instruct-2507-GGUF · Q5_K_M | Fits | 84.533.8–93.0 | 256K | 30.5B | p85 | – | |
| google/gemma-4-31B-it unsloth/gemma-4-31B-it-GGUF · Q5_K_M | Fits | 19.57.80–21.4 | 41K | 31.3B | p63 | – | |
| Qwen/Qwen3.5-27B unsloth/Qwen3.5-27B-GGUF · Q6_K | Fits | 22.69.00–24.9 | 43K | 27.8B | p60 | p46 | |
| Qwen/Qwen3.8-27B unsloth/Qwen3.8-27B-GGUF · Q4_1 | Fits | 27.811.1–30.6 | 111K | 27.8B | p59 | p33 | |
| Qwen/Qwen3.6-27B unsloth/Qwen3.6-27B-GGUF · Q6_K | Fits | 22.69.00–24.8 | 42K | 27.8B | p51 | p54 | |
| Qwen/Qwen3.5-35B-A3B unsloth/Qwen3.5-35B-A3B-GGUF · Q4_K_M | Fits | 12951.5–142 | 256K | 36B | p49 | p35 | |
| google/gemma-4-26B-A4B-it unsloth/gemma-4-26B-A4B-it-GGUF · MXFP4_MOE | Fits | 61.824.7–68.0 | 256K | 25.8B | p49 | – | |
| Qwen/Qwen3.5-9B unsloth/Qwen3.5-9B-GGUF · Q8_0 | Fits | 49.619.8–54.5 | 256K | 9.7B | p46 | – | |
| ornith-ai/Ornith-1.5-9B BF16 | Fits | 28.511.4–31.3 | 149K | 9.7B | p46 | p24 | |
| Qwen/Qwen3.6-35B-A3B unsloth/Qwen3.6-35B-A3B-MTP-GGUF · MXFP4_MOE | Fits | 12851.3–141 | 256K | 36B | p43 | p41 | |
| openbmb/MiniCPM-V-4.7-35B-A3B openbmb/MiniCPM-V-4.7-35B-A3B-gguf · Q4_K_M | Fits | 13152.2–144 | 256K | 35.2B | p42 | – | |
| ornith-ai/Ornith-1.0-35B bartowski/deepreinforce-ai_Ornith-1.0-35B-GGUF · Q4_K_L | Fits | 12951.7–142 | 256K | 35.1B | – | p40 | |
| openai/gpt-oss-20b unsloth/gpt-oss-20b-GGUF · F16 | Fits | 10943.7–120 | 128K | 21.1B | p39 | p15 | |
| Qwen/Qwen3.5-4B BF16 | Fits | 46.918.8–51.6 | 256K | 4.7B | p39 | – | |
| google/gemma-4-12B-it unsloth/gemma-4-12b-it-GGUF · Q8_0 | Fits | 34.713.9–38.2 | 256K | 12B | p38 | – | |
| Qwen/Qwen3-4B-Instruct-2507 BF16 | Fits | 44.217.7–48.6 | 108K | 4B | p38 | – | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 bartowski/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF · Q4_K_S | Fits | 11947.6–131 | 256K | 32.9B | p37 | p9 | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 MIXED_PRECISION | Fits | 12148.5–133 | 1M | 47.1B | p36 | p9 | |
| nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 unsloth/Nemotron-3-Nano-30B-A3B-GGUF · Q5_K_S | Fits | 11847.0–129 | 256K | 31.6B | p33 | – | |
| poolside/Laguna-XS-2.1 poolside/Laguna-XS-2.1-GGUF · Q4_K_M | Fits | 13051.9–143 | 256K | 33.4B | – | p30 | |
| Qwen/Qwen2.5-7B BF16 | Fits | 32.413.0–35.7 | 128K | 7.6B | p29 | – | |
| google/gemma-4-E4B-it unsloth/gemma-4-E4B-it-GGUF · Q8_0 | Fits | 62.625.0–68.9 | 128K | 8B | p27 | – | |
| openbmb/MiniCPM5-2B BF16 | Fits | 70.628.2–77.6 | 128K | 2.5B | p27 | p3 | |
| poolside/Laguna-XS.2 Lucebox/Laguna-XS.2-GGUF · Q4_K_M | Fits | 13051.9–143 | 256K | 33.4B | – | p25 | |
| Qwen/Qwen3.5-2B BF16 | Fits | 80.632.2–88.6 | 256K | 2.3B | p24 | – | |
| ornith-ai/Ornith-1.0-9B unsloth/Ornith-1.0-9B-GGUF · Q8_0 | Fits | 49.719.9–54.7 | 256K | 9.4B | – | p20 | |
| google/gemma-4-E2B-it BF16 | Fits | 67.226.9–74.0 | 128K | 5.1B | p18 | – | |
| Qwen/Qwen2.5-1.5B BF16 | Fits | 93.137.2–102 | 128K | 1.5B | p15 | – | |
| mistralai/Mistral-7B-Instruct-v0.2 BF16 | Fits | 30.412.2–33.5 | 32K | 7.2B | p14 | – | |
| Qwen/Qwen2.5-0.5B BF16 | Fits | 15863.0–173 | 32K | 0.5B | p5 | – | |
| Qwen/Qwen3.5-0.8B BF16 | Fits | 13353.0–146 | 256K | 0.9B | p4 | – | |
| incoai/Qwen3.8-27B-DFlash2 BF16 | Fits | 15863.0–173 | 256K | 1.9B | – | – | |
| Qwen/Qwen2.5-0.5B-Instruct BF16 | Fits | 15863.0–173 | 32K | 0.5B | – | – | |
| datalab-to/surya-ocr-2 BF16 | Fits | 14558.0–160 | 256K | 0.7B | – | – | |
| Qwen/Qwen3.5-0.8B-Base BF16 | Fits | 13353.0–146 | 256K | 0.9B | – | – | |
| baidu/Unlimited-OCR-2 AtomicChat/Unlimited-OCR-2-GGUF · BF16 | Fits | 12750.9–140 | 128K | 3.3B | – | – | |
| ornith-ai/Ornith-1.0-35B-GGUF Q4_K_M | Fits | 11746.9–129 | 256K | 34.7B | – | – | |
| ornith-ai/Ornith-1.5-35B-A3B-GGUF Q4_K_M | Fits | 11646.5–128 | 256K | 35.5B | – | – | |
| openbmb/MiniCPM-V-4.7-1B BF16 | Fits | 11144.3–122 | 256K | 1.6B | – | – | |
| Qwen/Qwen3-0.6B-Base BF16 | Fits | 10542.0–116 | 32K | 0.6B | – | – | |
| Qwen/Qwen3-0.6B BF16 | Fits | 99.439.8–109 | 40K | 0.8B | – | – | |
| openbmb/MiniCPM5-2B-GGUF Q8_0 | Fits | 95.138.0–105 | 128K | 2.5B | – | – | |
| Qwen/Qwen2.5-1.5B-Instruct BF16 | Fits | 93.137.2–102 | 32K | 1.5B | – | – | |
| nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16 unsloth/NVIDIA-Nemotron-3-Nano-4B-GGUF · Q8_0 | Fits | 89.936.0–98.9 | 256K | 4B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct-FP8 ijohn07/Qwen3-Coder-30B-A3B-Instruct-FP8-Q4_K_M-GGUF · Q4_K_M | Fits | 89.235.7–98.1 | 229K | 30.5B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF · Q5_K_M | Fits | 84.533.8–93.0 | 256K | 30.5B | – | – | |
| Qwen/Qwen3-30B-A3B unsloth/Qwen3-30B-A3B-GGUF · Q5_K_M | Fits | 84.533.8–93.0 | 40K | 30.5B | – | – | |
| Qwen/Qwen2.5-3B-Instruct Qwen/Qwen2.5-3B-Instruct-GGUF · Q8_0 | Fits | 81.332.5–89.4 | 32K | 3.1B | – | – | |
| Qwen/Qwen2-VL-2B-Instruct BF16 | Fits | 76.630.6–84.3 | 32K | 2.2B | – | – | |
| Qwen/Qwen3-1.7B-Base BF16 | Fits | 74.529.8–81.9 | 32K | 1.7B | – | – | |
| Qwen/Qwen3-VL-2B-Instruct Qwen/Qwen3-VL-2B-Instruct-GGUF · F16 | Fits | 74.429.8–81.9 | 178K | 2.1B | – | – | |
| baidu/Unlimited-OCR BF16 | Fits | 72.929.2–80.2 | 32K | 3.3B | – | – | |
| Qwen/Qwen3-1.7B BF16 | Fits | 68.927.6–75.8 | 40K | 2B | – | – | |
| prism-ml/Bonsai-27B-mlx-1bit MLX-1BIT | Fits | 66.726.7–73.4 | 256K | 26.4B | – | – | |
| Qwen/Qwen2.5-VL-3B-Instruct unsloth/Qwen2.5-VL-3B-Instruct-GGUF · BF16 | Fits | 59.723.9–65.7 | 125K | 3.8B | – | – | |
| ornith-ai/Ornith-1.5-9B-NVFP4 MIXED_PRECISION | Fits | 53.421.4–58.7 | 256K | 9B | – | – | |
| Qwen/Qwen2.5-7B-Instruct Qwen/Qwen2.5-7B-Instruct-GGUF · Q8_0 | Fits | 51.720.7–56.8 | 32K | 7.6B | – | – | |
| Qwen/Qwen2.5-VL-7B-Instruct unsloth/Qwen2.5-VL-7B-Instruct-GGUF · Q8_0 | Fits | 51.520.6–56.6 | 125K | 8.3B | – | – | |
| Qwen/Qwen2-VL-7B-Instruct ggml-org/Qwen2-VL-7B-Instruct-GGUF · Q8_0 | Fits | 51.520.6–56.6 | 32K | 8.3B | – | – | |
| ornith-ai/Ornith-1.0-9B-GGUF Q8_0 | Fits | 49.920.0–54.9 | 256K | 9B | – | – | |
| ornith-ai/Ornith-1.5-9B-GGUF Q8_0 | Fits | 48.819.5–53.7 | 256K | 9.2B | – | – | |
| prism-ml/Ternary-Bonsai-27B-mlx-2bit MLX-2BIT | Fits | 48.719.5–53.5 | 232K | 26.4B | – | – | |
| Qwen/Qwen3.5-4B-Base BF16 | Fits | 46.918.8–51.6 | 256K | 4.7B | – | – | |
| Qwen/Qwen2.5-Coder-7B-Instruct Qwen/Qwen2.5-Coder-7B-Instruct-GGUF · Q4_K_M | Fits | 46.718.7–51.4 | 32K | 7.6B | – | – | |
| HauhauCS/Gemma-4-E4B-Uncensored-HauhauCS-Aggressive Q8_K_P | Fits | 46.518.6–51.2 | 128K | 7.5B | – | – | |
| Qwen/Qwen3-8B-Base Qwen/Qwen3-Embedding-8B-GGUF · Q8_0 | Fits | 46.318.5–50.9 | 32K | 8.2B | – | – | |
| Qwen/Qwen3-4B BF16 | Fits | 44.217.7–48.6 | 40K | 4B | – | – | |
| Qwen/Qwen3-8B unsloth/Qwen3-8B-GGUF · Q8_0 | Fits | 44.117.7–48.6 | 40K | 8.2B | – | – | |
| Qwen/Qwen3-VL-4B-Instruct Qwen/Qwen3-VL-4B-Instruct-GGUF · F16 | Fits | 44.117.7–48.5 | 108K | 4.4B | – | – | |
| Qwen/Qwen3-VL-8B-Instruct Qwen/Qwen3-VL-8B-Instruct-GGUF · Q8_0 | Fits | 44.017.6–48.4 | 108K | 8.8B | – | – | |
| datalab-to/chandra-ocr-2 BF16 | Fits | 42.517.0–46.7 | 256K | 5.3B | – | – | |
| Qwen/Qwen3-14B unsloth/Qwen3-14B-GGUF · Q8_0 | Fits | 28.411.4–31.3 | 40K | 14.8B | – | – | |
| Qwen/Qwen2.5-14B-Instruct Qwen/Qwen2.5-14B-Instruct-GGUF · Q8_0 | Fits | 27.611.0–30.4 | 32K | 14.8B | – | – | |
| Qwen/Qwen2.5-Coder-14B-Instruct Qwen/Qwen2.5-Coder-14B-Instruct-GGUF · Q4_K_M | Fits | 25.010.0–27.5 | 32K | 14.8B | – | – | |
| RadixArk/Qwen3.8-27B-NVFP4 MIXED_PRECISION | Fits | 23.49.40–25.7 | 35K | 26.4B | – | – | |
| Qwen/Qwen2.5-32B-Instruct bartowski/Qwen2.5-32B-Instruct-GGUF · Q4_K_M | Fits | 21.58.60–23.7 | 32K | 32.8B | – | – | |
| Qwen/Qwen2.5-Coder-32B-Instruct bartowski/Qwen2.5-Coder-32B-Instruct-GGUF · Q4_K_M | Fits | 21.58.60–23.7 | 32K | 32.8B | – | – | |
| Qwen/Qwen3-32B unsloth/Qwen3-32B-GGUF · Q4_K_M | Fits | 21.38.50–23.4 | 35K | 32.8B | – | – | |
| farbodtavakkoli/OTel-2.0-LLM-31B-IT wfakhri/OTel-2.0-LLM-31B-IT-GGUF · Q5_K_M | Fits | 19.37.70–21.3 | 38K | 31.3B | – | – | |
| nvidia/Qwen3.6-35B-A3B-NVFP4 MIXED_PRECISION | Tight fit | 12650.6–139 | 256K | 34B | – | – | |
| ornith-ai/Ornith-1.5-35B-A3B-NVFP4 MIXED_PRECISION | Tight fit | 12650.6–139 | 256K | 34B | – | – | |
| zai-org/GLM-5.3 unsloth/GLM-5.3-GGUF · BF16 | Offload | 1.900.90–2.80 | 0 | 753.3B | p81 | p58 | |
| Qwen/Qwen3.5-122B-A10B unsloth/Qwen3.5-122B-A10B-GGUF · Q4_K_M | Offload | 17.67.00–26.4 | 256K | 125.1B | p77 | p49 | |
| zai-org/GLM-5 unsloth/GLM-5-GGUF · BF16 | Offload | 1.900.90–2.80 | 0 | 753.9B | p74 | p71 | |
| zai-org/GLM-5.1 unsloth/GLM-5.1-GGUF · BF16 | Offload | 1.900.90–2.80 | 0 | 753.9B | p53 | p68 | |
| Qwen/Qwen3.8-Flash-Next ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF · IQ3_XXS | Offload | 26.410.5–39.5 | 256K | 180B | p68 | p51 | |
| thinkingmachines/Inkling unsloth/inkling-GGUF · Q8_0 | Offload | 8.404.20–12.6 | 0 | 952.4B | p68 | p45 | |
| MiniMaxAI/MiniMax-M2.5 unsloth/MiniMax-M2.5-GGUF · UD-IQ2_M | Offload | 15.07.50–22.6 | 45K | 228.7B | p46 | p61 | |
| deepseek-ai/DeepSeek-V3.2 unsloth/DeepSeek-V3.2-GGUF · BF16 | Offload | 1.800.90–2.70 | 0 | 685.4B | p60 | p34 | |
| nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16 unsloth/NVIDIA-Nemotron-3-Super-120B-A12B-GGUF · MXFP4_MOE | Offload | 13.05.20–19.5 | 256K | 123.6B | p50 | p15 | |
| openai/gpt-oss-120b unsloth/gpt-oss-120b-GGUF · F16 | Offload | 22.311.2–33.5 | 128K | 117B | p47 | p15 | |
| poolside/Laguna-S-2.1 Myric/Laguna-S-2.1-APEX-GGUF · V2 [quality] | Offload | 15.26.10–22.7 | 204K | 117.6B | – | p34 | |
| ornith-ai/Ornith-1.5-35B-A3B-FP8 COMPRESSED-TENSORS | Offload | 38.215.3–57.3 | 256K | 36B | – | – | |
| ornith-ai/Ornith-1.5-122B-GGUF Q4_K_M | Offload | 17.06.80–25.6 | 256K | 124.6B | – | – |
Quality = benchmark percentile against all indexed models (p50 = median). Speeds are single-user decode ranges.