Compatibility Matrix

Best community run per VRAM tier for the most-run open-weight models. Cells show tok/s and quantization — the same model often lands in multiple tiers via different quants. Full board · Submit your run.

Model familyAll Arcee Baichuan BTL Cohere Deepseek Deepseek-Coder Deepseek-R1 DeepSeek-V4.1-Flash Gemma GLM Gpt Hunyuan Kimi Kwaipilot Laguna LFM Ling Llama MiniCPM-V Minimax Mistral MobileLLM Muse Nemotron Ornith Phi Qwen Qwen3.8 RNJ Spark unknown

Model	Family	<24GB	24–48GB	>48GB	runs
mistralai_Mistral-Small-3.2-24B-Instruct-2506	Mistral	21.8 · Q4_K_M	—	—	1
Ministral-3-14B-Reasoning-2512	Mistral	—	89.4 · Q8_0	—	1
DeepHermes-3-Mistral-24B-Preview	Mistral	24.5 · Q4_K_M	—	—	1
Ministral-3-3B-Instruct-2512	Mistral	94.9 · Q4_K_M	—	—	1

4 runs · last run Aug 18, 2026. Filter by family above; full board for every row.

VRAM tiers: <24 GB · 24–48 GB · >48 GB. A cell shows the best single-stream tok/s and quant recorded in that tier — not a guarantee for every loadout. Imported rows carry their original methodology link.

Related guides: best GPU for Qwen3 · quantization · Strix Halo 128GB