GraySoft
Projects Models Compare Cloud benchmarks FAQ Download guIDE →

Models for 32 GB+ VRAM

Large local models and high-bit quants for 32 GB+ VRAM setups. Sorted by Hugging Face downloads. Smallest GGUF file size shown per model.

AbteeXAILab/lumynax-reasoning-deepseek-r1-distill-llama-70b-gguf
AbteeXAILab · from 39.60 GB · 626 downloads
mradermacher/MiniMax-M2.1-REAP-30-GGUF
mradermacher · from 55.04 GB · 96 downloads
zTrojan/Qwen3.5-122B-A10B-REAP20-APEX-GGUF
zTrojan · from 34.63 GB · 1 downloads
Farfuad77/Qwen3.5-397B-A17B-Opus-4.6-Reasoning-Uncensored-GGUF
Farfuad77 · from 134.80 GB · 0 downloads
forkjoin-ai/qwen2.5-72b-instruct-gguf
forkjoin-ai · from 44.16 GB · 0 downloads
zTrojan/Qwen3.5-122B-A10B-REAP30-APEX-GGUF
zTrojan · from 30.75 GB · 0 downloads

Run models locally with guIDE

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.

Download guIDE → · Browse 524k+ models · Compare models