| 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 | 14.45.80–18.0 | 33K | 30.5B | p85 | – | |
| google/gemma-4-31B-it unsloth/gemma-4-31B-it-GGUF · Q5_K_M | Fits | 5.904.70–7.30 | 43K | 31.3B | p62 | – | |
| Qwen/Qwen3.5-27B unsloth/Qwen3.5-27B-GGUF · Q6_K | Fits | 7.202.90–7.90 | 43K | 27.8B | p60 | p46 | |
| Qwen/Qwen3.8-27B unsloth/Qwen3.8-27B-GGUF · Q4_1 | Fits | 8.803.50–9.70 | 115K | 27.8B | p58 | p31 | |
| Qwen/Qwen3.6-27B unsloth/Qwen3.6-27B-GGUF · Q6_K | Fits | 7.102.90–7.90 | 42K | 27.8B | p50 | p54 | |
| Qwen/Qwen3.5-35B-A3B unsloth/Qwen3.5-35B-A3B-GGUF · Q4_K_M | Fits | 27.511.0–34.3 | 134K | 36B | p48 | p35 | |
| google/gemma-4-26B-A4B-it unsloth/gemma-4-26B-A4B-it-GGUF · MXFP4_MOE | Fits | 21.88.70–23.9 | 256K | 25.8B | p48 | – | |
| Qwen/Qwen3.5-9B unsloth/Qwen3.5-9B-GGUF · Q8_0 | Fits | 17.16.90–18.8 | 256K | 9.7B | p46 | – | |
| ornith-ai/Ornith-1.5-9B AtomicChat/Ornith-1.5-9B-GGUF · Q8_0 | Fits | 17.16.90–18.8 | 256K | 9.7B | p44 | p23 | |
| Qwen/Qwen3.6-35B-A3B unsloth/Qwen3.6-35B-A3B-MTP-GGUF · MXFP4_MOE | Fits | 27.411.0–34.2 | 126K | 36B | p43 | p41 | |
| openbmb/MiniCPM-V-4.7-35B-A3B openbmb/MiniCPM-V-4.7-35B-A3B-gguf · Q5_K_S | Fits | 26.610.6–33.2 | 45K | 35.2B | p42 | – | |
| openai/gpt-oss-20b unsloth/gpt-oss-20b-GGUF · F16 | Fits | 26.010.4–32.5 | 128K | 21.1B | p39 | p15 | |
| Qwen/Qwen3.5-4B unsloth/Qwen3.5-4B-GGUF · Q8_0 | Fits | 28.111.3–30.9 | 256K | 4.7B | p39 | – | |
| ornith-ai/Ornith-1.0-35B bartowski/deepreinforce-ai_Ornith-1.0-35B-GGUF · Q5_K_S | Fits | 26.510.6–33.1 | 37K | 35.1B | – | p38 | |
| google/gemma-4-12B-it unsloth/gemma-4-12b-it-GGUF · Q8_0 | Fits | 8.807.50–10.1 | 256K | 12B | p38 | – | |
| Qwen/Qwen3-4B-Instruct-2507 BF16 | Fits | 14.512.3–16.7 | 116K | 4B | p38 | – | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 bartowski/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF · Q5_K_S | Fits | 26.210.5–32.8 | 33K | 32.9B | p37 | p8 | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 MIXED_PRECISION | Fits | 47.619.0–52.3 | 449K | 47.1B | p36 | p8 | |
| nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 unsloth/Nemotron-3-Nano-30B-A3B-GGUF · Q4_K_M | Fits | 26.410.6–33.0 | 58K | 31.6B | p33 | – | |
| poolside/Laguna-XS-2.1 bartowski/Laguna-XS-2.1-GGUF · Q5_K_L | Fits | 29.511.8–36.9 | 34K | 33.4B | – | p30 | |
| poolside/Laguna-XS.2 Lucebox/Laguna-XS.2-GGUF · Q4_K_M | Fits | 31.912.7–39.8 | 109K | 33.4B | – | p25 | |
| ornith-ai/Ornith-1.0-9B unsloth/Ornith-1.0-9B-GGUF · Q8_0 | Fits | 17.26.90–18.9 | 256K | 9.4B | – | p19 | |
| ornith-ai/Ornith-1.5-35B-A3B-NVFP4 MIXED_PRECISION | Fits | 45.818.3–50.4 | 71K | 34B | – | – | |
| nvidia/Qwen3.6-35B-A3B-NVFP4 MIXED_PRECISION | Fits | 45.818.3–50.4 | 71K | 34B | – | – | |
| ornith-ai/Ornith-1.0-35B-GGUF Q4_K_M | Fits | 24.79.90–30.8 | 172K | 34.7B | – | – | |
| ornith-ai/Ornith-1.5-35B-A3B-GGUF Q4_K_M | Fits | 24.49.80–30.5 | 147K | 35.5B | – | – | |
| prism-ml/Bonsai-27B-mlx-1bit MLX-1BIT | Fits | 21.98.80–24.1 | 256K | 26.4B | – | – | |
| ornith-ai/Ornith-1.5-9B-GGUF Q8_0 | Fits | 16.86.70–18.5 | 256K | 9.2B | – | – | |
| prism-ml/Ternary-Bonsai-27B-mlx-2bit MLX-2BIT | Fits | 15.76.30–17.3 | 248K | 26.4B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct-FP8 ijohn07/Qwen3-Coder-30B-A3B-Instruct-FP8-Q4_K_M-GGUF · Q4_K_M | Fits | 15.16.00–18.8 | 64K | 30.5B | – | – | |
| Qwen/Qwen3-30B-A3B unsloth/Qwen3-30B-A3B-GGUF · Q5_K_M | Fits | 14.45.80–18.0 | 33K | 30.5B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF · Q5_K_M | Fits | 14.45.80–18.0 | 33K | 30.5B | – | – | |
| Qwen/Qwen3-14B unsloth/Qwen3-14B-GGUF · Q8_0 | Fits | 8.006.40–10.0 | 40K | 14.8B | – | – | |
| Qwen/Qwen2.5-14B-Instruct Qwen/Qwen2.5-14B-Instruct-GGUF · Q8_0 | Fits | 7.706.10–9.60 | 32K | 14.8B | – | – | |
| RadixArk/Qwen3.8-27B-NVFP4 MIXED_PRECISION | Fits | 7.403.00–8.10 | 51K | 26.4B | – | – | |
| Qwen/Qwen2.5-Coder-14B-Instruct Qwen/Qwen2.5-Coder-14B-Instruct-GGUF · Q4_K_M | Fits | 7.005.60–8.70 | 32K | 14.8B | – | – | |
| Qwen/Qwen2.5-32B-Instruct Qwen/Qwen2.5-32B-Instruct-GGUF · Q4_K_M | Fits | 5.904.70–7.40 | 32K | 32.8B | – | – | |
| farbodtavakkoli/OTel-2.0-LLM-31B-IT wfakhri/OTel-2.0-LLM-31B-IT-GGUF · Q5_K_M | Fits | 5.804.60–7.30 | 39K | 31.3B | – | – | |
| Qwen/Qwen2.5-Coder-32B-Instruct bartowski/Qwen2.5-Coder-32B-Instruct-GGUF · Q4_K_L | Fits | 5.804.60–7.20 | 32K | 32.8B | – | – | |
| Qwen/Qwen3-32B unsloth/Qwen3-32B-GGUF · Q4_K_M | Fits | 5.604.50–7.00 | 40K | 32.8B | – | – |
Quality = benchmark percentile against all indexed models (p50 = median). Speeds are single-user decode ranges.