| Model | Verdict | Decode tok/s | Max ctx | Size | Chat | Coding | |
|---|---|---|---|---|---|---|---|
| Qwen/Qwen3.5-9B unsloth/Qwen3.5-9B-GGUF · Q6_K | Fits | 7.302.90–8.00 | 38K | 9.7B | p46 | – | |
| ornith-ai/Ornith-1.5-9B AtomicChat/Ornith-1.5-9B-GGUF · Q6_K | Fits | 7.402.90–8.10 | 42K | 9.7B | p46 | p24 | |
| Qwen/Qwen3.5-4B unsloth/Qwen3.5-4B-GGUF · Q8_0 | Fits | 10.44.10–11.4 | 90K | 4.7B | p39 | – | |
| google/gemma-4-12B-it unsloth/gemma-4-12b-it-GGUF · Q4_K_M | Fits | 6.302.50–7.00 | 35K | 12B | p38 | – | |
| Qwen/Qwen3-4B-Instruct-2507 unsloth/Qwen3-4B-Instruct-2507-GGUF · Q8_0 | Fits | 7.603.10–8.40 | 43K | 4B | p38 | – | |
| ornith-ai/Ornith-1.0-9B unsloth/Ornith-1.0-9B-GGUF · Q6_K | Fits | 7.302.90–8.00 | 38K | 9.4B | – | p20 | |
| prism-ml/Bonsai-27B-mlx-1bit MLX-1BIT | Fits | 8.603.40–9.40 | 44K | 26.4B | – | – | |
| ornith-ai/Ornith-1.5-9B-GGUF Q6_K | Fits | 7.202.90–7.90 | 34K | 9.2B | – | – | |
| Qwen/Qwen2.5-Coder-14B-Instruct bartowski/Qwen2.5-Coder-14B-Instruct-GGUF · IQ2_XS | Fits | 6.602.60–7.20 | 32K | 14.8B | – | – | |
| poolside/Laguna-XS.2 poolside-laguna-hackathon/Piscina-XS.2-GGUF · IQ1_M | Tight fit | 37.815.1–41.6 | 5K | 33.4B | – | p25 | |
| Qwen/Qwen3-30B-A3B-Instruct-2507 MaziyarPanahi/Qwen3-30B-A3B-Instruct-2507-GGUF · Q6_K | Unknown | – | – | 30.5B | p85 | – |
Quality = benchmark percentile against all indexed models (p50 = median). Speeds are single-user decode ranges.