// Open-model VRAM table · family
Baichuan: local deployment & VRAM
2 open-weight models from Baichuan (7 official repos incl. quantised builds). At 4-bit they range from 24.5 GB (Baichuan-M2-32B) to 159 GB (Baichuan-M3-235B).
official orgs: baichuan-inclatest 2026-01-13memory = estimate
01Text LLM (2)
| Model | Params | Context | BF16 | 4-bit | Smallest GPU (4-bit) | Mac (4-bit) | Licence | Released |
|---|---|---|---|---|---|---|---|---|
| Baichuan-M3-235B | 235.1B | 40K | 519 GB | 159 GB | 4 × RTX PRO 6000 (384GB) | Mac 256GB | Commercial OK | 2026-01-13 |
| Baichuan-M2-32B ↗ | 32.8B | 128K | 74.7 GB | 24.5 GB | RTX 5090 32GB | Mac 64GB | Commercial OK | 2025-08-10 |
02Other families
Qwen · IBM Granite · NVIDIA Nemotron · Gemma · Mistral / Magistral / Devstral · GLM / CogVideoX (Z.ai) · DeepSeek · Ai2 OLMo / Molmo · Ant Ling / Ring / Ming · Liquid LFM · MiniCPM / VoxCPM · Tencent Hunyuan / Hy · Phi · Llama · Meituan LongCat · Jina embeddings & rerankers · Wan · Kimi · Cohere Command / Aya · Xiaomi MiMo · FLUX · Microsoft VibeVoice / Fara / Harrier · InternLM / Intern-S · StepFun Step · MiniMax · Baidu ERNIE · NVIDIA Parakeet / Canary speech · Stable Diffusion / Stable Audio · Yi (01.AI) · ByteDance Seed · SmolLM / SmolVLM · BGE embeddings & rerankers · LTX Video · FunAudioLLM speech · gpt-oss · poolside Laguna · Whisper · Thinking Machines Inkling · xAI Grok · rednote dots
Totals include 10% + 0.5GB overhead and one 8K-token KV cache for text models. Estimates only. Method →