| Model | Verdict | Decode tok/s | Max ctx | Size | Chat | Coding | |
|---|---|---|---|---|---|---|---|
| Qwen/Qwen3-30B-A3B-Instruct-2507 unsloth/Qwen3-30B-A3B-Instruct-2507-GGUF · Q8_0 | Fits | 21.38.50–23.4 | 256K | 30.5B | p85 | – | |
| google/gemma-4-31B-it unsloth/gemma-4-31B-it-GGUF · Q8_0 | Fits | 3.101.20–3.40 | 256K | 31.3B | p63 | – | |
| Qwen/Qwen3.5-27B unsloth/Qwen3.5-27B-GGUF · Q8_0 | Fits | 3.801.50–4.20 | 256K | 27.8B | p60 | p46 | |
| Qwen/Qwen3.8-27B unsloth/Qwen3.8-27B-GGUF · Q8_0 | Fits | 3.701.50–4.10 | 256K | 27.8B | p59 | p33 | |
| Qwen/Qwen3.6-27B unsloth/Qwen3.6-27B-GGUF · Q8_0 | Fits | 3.801.50–4.20 | 256K | 27.8B | p51 | p54 | |
| Qwen/Qwen3.5-35B-A3B unsloth/Qwen3.5-35B-A3B-GGUF · Q8_0 | Fits | 34.313.7–37.7 | 256K | 36B | p49 | p35 | |
| google/gemma-4-26B-A4B-it unsloth/gemma-4-26B-A4B-it-GGUF · Q8_0 | Fits | 25.610.2–28.1 | 256K | 25.8B | p49 | – | |
| Qwen/Qwen3.5-9B unsloth/Qwen3.5-9B-GGUF · Q8_0 | Fits | 11.64.70–12.8 | 256K | 9.7B | p46 | – | |
| ornith-ai/Ornith-1.5-9B AtomicChat/Ornith-1.5-9B-GGUF · Q8_0 | Fits | 11.64.70–12.8 | 256K | 9.7B | p46 | p24 | |
| Qwen/Qwen3.6-35B-A3B unsloth/Qwen3.6-35B-A3B-MTP-GGUF · Q8_0 | Fits | 33.613.4–37.0 | 256K | 36B | p43 | p41 | |
| openbmb/MiniCPM-V-4.7-35B-A3B openbmb/MiniCPM-V-4.7-35B-A3B-gguf · Q8_0 | Fits | 34.413.8–37.8 | 256K | 35.2B | p42 | – | |
| ornith-ai/Ornith-1.0-35B unsloth/Ornith-1.0-35B-GGUF · Q8_0 | Fits | 34.413.8–37.8 | 256K | 35.1B | – | p40 | |
| openai/gpt-oss-20b MXFP4 | Fits | 19.57.80–21.4 | 128K | 21.1B | p39 | p15 | |
| Qwen/Qwen3.5-4B unsloth/Qwen3.5-4B-GGUF · Q8_0 | Fits | 20.68.20–22.7 | 256K | 4.7B | p39 | – | |
| google/gemma-4-12B-it unsloth/gemma-4-12b-it-GGUF · Q8_0 | Fits | 7.903.20–8.70 | 256K | 12B | p38 | – | |
| Qwen/Qwen3-4B-Instruct-2507 unsloth/Qwen3-4B-Instruct-2507-GGUF · Q8_0 | Fits | 15.66.20–17.1 | 256K | 4B | p38 | – | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 BF16 | Fits | 31.712.7–34.9 | 256K | 32.9B | p37 | p9 | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 MIXED_PRECISION | Fits | 43.217.3–47.6 | 1M | 47.1B | p36 | p9 | |
| nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 BF16 | Fits | 32.913.2–36.2 | 256K | 31.6B | p33 | – | |
| poolside/Laguna-XS-2.1 bartowski/Laguna-XS-2.1-GGUF · Q8_0 | Fits | 33.713.5–37.1 | 256K | 33.4B | – | p30 | |
| poolside/Laguna-XS.2 BF16 | Fits | 16.46.60–18.1 | 256K | 33.4B | – | p25 | |
| ornith-ai/Ornith-1.0-9B unsloth/Ornith-1.0-9B-GGUF · Q8_0 | Fits | 11.74.70–12.8 | 256K | 9.4B | – | p20 | |
| ornith-ai/Ornith-1.5-35B-A3B-NVFP4 MIXED_PRECISION | Fits | 49.119.6–54.0 | 256K | 34B | – | – | |
| nvidia/Qwen3.6-35B-A3B-NVFP4 MIXED_PRECISION | Fits | 49.019.6–53.9 | 256K | 34B | – | – | |
| ornith-ai/Ornith-1.0-35B-GGUF Q8_0 | Fits | 28.411.4–31.3 | 256K | 34.7B | – | – | |
| ornith-ai/Ornith-1.5-35B-A3B-GGUF Q8_0 | Fits | 27.911.2–30.7 | 256K | 35.5B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct-FP8 FP8 | Fits | 21.48.60–23.6 | 256K | 30.5B | – | – | |
| Qwen/Qwen3-30B-A3B unsloth/Qwen3-30B-A3B-GGUF · Q8_0 | Fits | 21.38.50–23.4 | 40K | 30.5B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF · Q8_0 | Fits | 21.38.50–23.4 | 256K | 30.5B | – | – | |
| prism-ml/Bonsai-27B-mlx-1bit MLX-1BIT | Fits | 17.46.90–19.1 | 256K | 26.4B | – | – | |
| prism-ml/Ternary-Bonsai-27B-mlx-2bit MLX-2BIT | Fits | 11.54.60–12.6 | 256K | 26.4B | – | – | |
| ornith-ai/Ornith-1.5-9B-GGUF Q8_0 | Fits | 11.44.60–12.6 | 256K | 9.2B | – | – | |
| RadixArk/Qwen3.8-27B-NVFP4 MIXED_PRECISION | Fits | 4.901.90–5.40 | 256K | 26.4B | – | – | |
| Qwen/Qwen3-32B unsloth/Qwen3-32B-GGUF · Q6_K | Fits | 3.501.40–3.90 | 40K | 32.8B | – | – | |
| Qwen/Qwen2.5-32B-Instruct Qwen/Qwen2.5-32B-Instruct-GGUF · Q6_K | Fits | 3.501.40–3.90 | 32K | 32.8B | – | – | |
| Qwen/Qwen3-14B BF16 | Fits | 3.501.40–3.80 | 40K | 14.8B | – | – | |
| Qwen/Qwen2.5-14B-Instruct BF16 | Fits | 3.401.40–3.80 | 32K | 14.8B | – | – | |
| Qwen/Qwen2.5-Coder-14B-Instruct BF16 | Fits | 3.401.40–3.80 | 32K | 14.8B | – | – | |
| farbodtavakkoli/OTel-2.0-LLM-31B-IT wfakhri/OTel-2.0-LLM-31B-IT-GGUF · Q8_0 | Fits | 3.101.20–3.40 | 256K | 31.3B | – | – | |
| Qwen/Qwen2.5-Coder-32B-Instruct Qwen/Qwen2.5-Coder-32B-Instruct-GGUF · Q3_K_M | Fits | 3.001.20–3.30 | 32K | 32.8B | – | – |
Quality = benchmark percentile against all indexed models (p50 = median). Speeds are single-user decode ranges.