| Model | Verdict | Decode tok/s | Max ctx | Size | Chat | Coding | |
|---|---|---|---|---|---|---|---|
| Qwen/Qwen3-30B-A3B-Instruct-2507 bartowski/Qwen_Qwen3-30B-A3B-Instruct-2507-GGUF · Q3_K_M | Fits | 63.554.0–73.1 | 256K | 30.5B | p85 | – | |
| google/gemma-4-31B-it unsloth/gemma-4-31B-it-GGUF · Q3_K_S | Fits | 31.526.8–36.3 | 39K | 31.3B | p63 | – | |
| Qwen/Qwen3.5-27B unsloth/Qwen3.5-27B-GGUF · IQ4_XS | Fits | 32.627.7–37.5 | 38K | 27.8B | p60 | p46 | |
| Qwen/Qwen3.8-27B unsloth/Qwen3.8-27B-GGUF · UD-IQ4_XS | Fits | 33.628.5–38.6 | 53K | 27.8B | p59 | p33 | |
| Qwen/Qwen3.6-27B unsloth/Qwen3.6-27B-GGUF · Q3_K_M | Fits | 34.929.6–40.1 | 40K | 27.8B | p51 | p54 | |
| Qwen/Qwen3.5-35B-A3B unsloth/Qwen3.5-35B-A3B-GGUF · Q3_K_S | Fits | 125107–144 | 256K | 36B | p49 | p35 | |
| google/gemma-4-26B-A4B-it unsloth/gemma-4-26B-A4B-it-GGUF · UD-IQ4_NL | Fits | 11496.6–131 | 256K | 25.8B | p49 | – | |
| Qwen/Qwen3.5-9B unsloth/Qwen3.5-9B-GGUF · Q8_0 | Fits | 60.151.1–69.1 | 215K | 9.7B | p46 | – | |
| ornith-ai/Ornith-1.5-9B AtomicChat/Ornith-1.5-9B-GGUF · Q8_0 | Fits | 60.151.1–69.1 | 215K | 9.7B | p46 | p24 | |
| Qwen/Qwen3.6-35B-A3B unsloth/Qwen3.6-35B-A3B-GGUF · UD-Q3_K_S | Fits | 125107–144 | 256K | 36B | p43 | p41 | |
| openbmb/MiniCPM-V-4.7-35B-A3B openbmb/MiniCPM-V-4.7-35B-A3B-gguf · IQ3_M | Fits | 125107–144 | 256K | 35.2B | p42 | – | |
| ornith-ai/Ornith-1.0-35B bartowski/deepreinforce-ai_Ornith-1.0-35B-GGUF · IQ3_XXS | Fits | 126108–145 | 256K | 35.1B | – | p40 | |
| openai/gpt-oss-20b unsloth/gpt-oss-20b-GGUF · F16 | Fits | 149127–171 | 128K | 21.1B | p39 | p15 | |
| Qwen/Qwen3.5-4B BF16 | Fits | 56.222.5–61.8 | 190K | 4.7B | p39 | – | |
| google/gemma-4-12B-it unsloth/gemma-4-12b-it-GGUF · Q8_0 | Fits | 30.926.3–35.6 | 155K | 12B | p38 | – | |
| Qwen/Qwen3-4B-Instruct-2507 BF16 | Fits | 51.120.5–56.3 | 52K | 4B | p38 | – | |
| poolside/Laguna-XS-2.1 bartowski/Laguna-XS-2.1-GGUF · Q3_K_S | Fits | 158134–182 | 256K | 33.4B | – | p30 | |
| poolside/Laguna-XS.2 poolside-laguna-hackathon/Piscina-XS.2-GGUF · IQ2 | Fits | 166141–191 | 256K | 33.4B | – | p25 | |
| ornith-ai/Ornith-1.0-9B unsloth/Ornith-1.0-9B-GGUF · Q8_0 | Fits | 60.351.2–69.3 | 215K | 9.4B | – | p20 | |
| Qwen/Qwen3-30B-A3B unsloth/Qwen3-30B-A3B-GGUF · Q3_K_S | Fits | 64.054.4–73.6 | 40K | 30.5B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF · Q3_K_S | Fits | 64.054.4–73.6 | 256K | 30.5B | – | – | |
| prism-ml/Bonsai-27B-mlx-1bit MLX-1BIT | Fits | 59.523.8–65.5 | 156K | 26.4B | – | – | |
| ornith-ai/Ornith-1.5-9B-GGUF Q8_0 | Fits | 59.050.1–67.8 | 208K | 9.2B | – | – | |
| prism-ml/Ternary-Bonsai-27B-mlx-2bit MLX-2BIT | Fits | 46.518.6–51.2 | 107K | 26.4B | – | – | |
| Qwen/Qwen3-14B unsloth/Qwen3-14B-GGUF · Q6_K | Fits | 37.832.1–43.5 | 40K | 14.8B | – | – | |
| Qwen/Qwen2.5-14B-Instruct Qwen/Qwen2.5-14B-Instruct-GGUF · Q6_K | Fits | 35.230.0–40.5 | 32K | 14.8B | – | – | |
| Qwen/Qwen2.5-Coder-14B-Instruct bartowski/Qwen2.5-Coder-14B-Instruct-GGUF · Q6_K_L | Fits | 34.629.4–39.8 | 32K | 14.8B | – | – | |
| farbodtavakkoli/OTel-2.0-LLM-31B-IT wfakhri/OTel-2.0-LLM-31B-IT-GGUF · IQ3_XS | Fits | 31.727.0–36.5 | 42K | 31.3B | – | – | |
| Qwen/Qwen2.5-Coder-32B-Instruct bartowski/Qwen2.5-Coder-32B-Instruct-GGUF · IQ2_M | Fits | 31.727.0–36.5 | 32K | 32.8B | – | – | |
| Qwen/Qwen2.5-32B-Instruct bartowski/Qwen2.5-32B-Instruct-GGUF · IQ2_M | Fits | 31.727.0–36.5 | 32K | 32.8B | – | – | |
| Qwen/Qwen3-32B unsloth/Qwen3-32B-GGUF · UD-IQ2_XXS | Fits | 30.425.9–35.0 | 40K | 32.8B | – | – | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 unsloth/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF · Q8_0 | Offload | 64.132.1–96.2 | 256K | 32.9B | p37 | p9 | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 MIXED_PRECISION | Offload | 98.439.4–148 | 1M | 47.1B | p36 | p9 | |
| nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 unsloth/Nemotron-3-Nano-30B-A3B-GGUF · Q8_0 | Offload | 66.033.0–99.0 | 256K | 31.6B | p33 | – | |
| ornith-ai/Ornith-1.5-35B-A3B-NVFP4 MIXED_PRECISION | Offload | 97.639.0–146 | 256K | 34B | – | – | |
| nvidia/Qwen3.6-35B-A3B-NVFP4 MIXED_PRECISION | Offload | 97.639.0–146 | 256K | 34B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct-FP8 FP8 | Offload | 40.416.1–60.5 | 139K | 30.5B | – | – |
Quality = benchmark percentile against all indexed models (p50 = median). Speeds are single-user decode ranges.