| Model | Verdict | Decode tok/s | Max ctx | Size | Chat | Coding | |
|---|---|---|---|---|---|---|---|
| Qwen/Qwen3-30B-A3B-Instruct-2507 unsloth/Qwen3-30B-A3B-Instruct-2507-GGUF · Q6_K | Fits | 46.018.4–50.6 | 43K | 30.5B | p85 | – | |
| google/gemma-4-31B-it unsloth/gemma-4-31B-it-GGUF · Q6_K | Fits | 11.810.0–13.6 | 54K | 31.3B | p63 | – | |
| Qwen/Qwen3.5-27B unsloth/Qwen3.5-27B-GGUF · Q8_0 | Fits | 11.74.70–12.9 | 32K | 27.8B | p60 | p46 | |
| Qwen/Qwen3.8-27B lmstudio-community/Qwen3.8-27B-MLX-6bit · MLX-6BIT | Fits | 14.35.70–15.7 | 103K | 27.8B | p59 | p33 | |
| Qwen/Qwen3.6-27B unsloth/Qwen3.6-27B-GGUF · Q8_0 | Fits | 11.74.70–12.9 | 32K | 27.8B | p51 | p54 | |
| Qwen/Qwen3.5-35B-A3B unsloth/Qwen3.5-35B-A3B-GGUF · Q6_K | Fits | 74.029.6–81.4 | 41K | 36B | p49 | p35 | |
| google/gemma-4-26B-A4B-it lmstudio-community/gemma-4-26B-A4B-it-MLX-8bit · MLX-8BIT | Fits | 54.321.7–59.7 | 55K | 25.8B | p49 | – | |
| Qwen/Qwen3.5-9B lmstudio-community/Qwen3.5-9B-MLX-8bit · MLX-8BIT | Fits | 30.812.3–33.8 | 256K | 9.7B | p46 | – | |
| ornith-ai/Ornith-1.5-9B AtomicChat/Ornith-1.5-9B-GGUF · Q8_0 | Fits | 33.013.2–36.3 | 256K | 9.7B | p46 | p24 | |
| Qwen/Qwen3.6-35B-A3B unsloth/Qwen3.6-35B-A3B-MTP-GGUF · MXFP4_MOE | Fits | 81.832.7–90.0 | 256K | 36B | p43 | p41 | |
| openbmb/MiniCPM-V-4.7-35B-A3B openbmb/MiniCPM-V-4.7-35B-A3B-gguf · Q6_K | Fits | 74.629.9–82.1 | 38K | 35.2B | p42 | – | |
| ornith-ai/Ornith-1.0-35B bartowski/deepreinforce-ai_Ornith-1.0-35B-GGUF · Q5_K_L | Fits | 78.231.3–86.0 | 181K | 35.1B | – | p40 | |
| openai/gpt-oss-20b MXFP4 | Fits | 50.320.1–55.4 | 72K | 21.1B | p39 | p15 | |
| Qwen/Qwen3.5-4B unsloth/Qwen3.5-4B-GGUF · Q8_0 | Fits | 51.920.8–57.1 | 256K | 4.7B | p39 | – | |
| google/gemma-4-12B-it unsloth/gemma-4-12b-it-GGUF · Q8_0 | Fits | 20.317.3–23.4 | 256K | 12B | p38 | – | |
| Qwen/Qwen3-4B-Instruct-2507 BF16 | Fits | 28.224.0–32.5 | 144K | 4B | p38 | – | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 bartowski/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF · Q5_K_L | Fits | 74.029.6–81.4 | 256K | 32.9B | p37 | p9 | |
| nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4 MIXED_PRECISION | Fits | 81.732.7–89.8 | 1013K | 47.1B | p36 | p9 | |
| nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 unsloth/Nemotron-3-Nano-30B-A3B-GGUF · Q5_K_M | Fits | 75.630.2–83.1 | 256K | 31.6B | p33 | – | |
| poolside/Laguna-XS-2.1 bartowski/Laguna-XS-2.1-GGUF · Q5_K_L | Fits | 85.234.1–93.7 | 122K | 33.4B | – | p30 | |
| poolside/Laguna-XS.2 Lucebox/Laguna-XS.2-GGUF · Q4_K_M | Fits | 90.236.1–99.3 | 211K | 33.4B | – | p25 | |
| ornith-ai/Ornith-1.0-9B lmstudio-community/Ornith-1.0-9B-MLX-8bit · MLX-8BIT | Fits | 30.812.3–33.9 | 256K | 9.4B | – | p20 | |
| ornith-ai/Ornith-1.5-35B-A3B-NVFP4 MIXED_PRECISION | Fits | 80.632.2–88.7 | 256K | 34B | – | – | |
| nvidia/Qwen3.6-35B-A3B-NVFP4 MIXED_PRECISION | Fits | 80.632.2–88.7 | 256K | 34B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct-FP8 tp7030/Qwen3-Coder-30B-A3B-Instruct-FP8-Q6_K-GGUF · Q6_K | Fits | 46.018.4–50.6 | 43K | 30.5B | – | – | |
| Qwen/Qwen3-30B-A3B unsloth/Qwen3-30B-A3B-GGUF · Q6_K | Fits | 46.018.4–50.6 | 40K | 30.5B | – | – | |
| Qwen/Qwen3-Coder-30B-A3B-Instruct unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF · Q6_K | Fits | 46.018.4–50.6 | 43K | 30.5B | – | – | |
| prism-ml/Bonsai-27B-mlx-1bit MLX-1BIT | Fits | 41.916.7–46.1 | 256K | 26.4B | – | – | |
| ornith-ai/Ornith-1.5-9B-GGUF Q8_0 | Fits | 32.513.0–35.7 | 256K | 9.2B | – | – | |
| prism-ml/Ternary-Bonsai-27B-mlx-2bit MLX-2BIT | Fits | 30.712.3–33.7 | 256K | 26.4B | – | – | |
| Qwen/Qwen3-14B lmstudio-community/Qwen3-14B-MLX-8bit · MLX-8BIT | Fits | 18.07.20–19.8 | 40K | 14.8B | – | – | |
| Qwen/Qwen2.5-14B-Instruct Qwen/Qwen2.5-14B-Instruct-GGUF · Q8_0 | Fits | 17.314.7–19.9 | 32K | 14.8B | – | – | |
| Qwen/Qwen2.5-Coder-14B-Instruct Qwen/Qwen2.5-Coder-14B-Instruct-GGUF · Q4_K_M | Fits | 15.713.3–18.1 | 32K | 14.8B | – | – | |
| RadixArk/Qwen3.8-27B-NVFP4 MIXED_PRECISION | Fits | 14.85.90–16.3 | 115K | 26.4B | – | – | |
| Qwen/Qwen2.5-32B-Instruct Qwen/Qwen2.5-32B-Instruct-GGUF · Q5_K_M | Fits | 11.910.1–13.7 | 32K | 32.8B | – | – | |
| farbodtavakkoli/OTel-2.0-LLM-31B-IT wfakhri/OTel-2.0-LLM-31B-IT-GGUF · Q6_K | Fits | 11.810.0–13.6 | 54K | 31.3B | – | – | |
| Qwen/Qwen3-32B unsloth/Qwen3-32B-GGUF · Q5_K_M | Fits | 11.59.80–13.3 | 40K | 32.8B | – | – | |
| Qwen/Qwen2.5-Coder-32B-Instruct Qwen/Qwen2.5-Coder-32B-Instruct-GGUF · Q2_K | Fits | 11.49.70–13.2 | 32K | 32.8B | – | – |
Quality = benchmark percentile against all indexed models (p50 = median). Speeds are single-user decode ranges.