PyPI page
Home page
Author:
Gavin Li
Summary:
AirLLM runs 70B large language models on a single 4GB GPU without quantization, distillation or pruning. Kimi K3 2.8T on under 4GB, Qwen3.8-Flash-Next 125B on 6GB, DeepSeek-V3 671B on ~12GB. Train Qwen3.8-Flash-Next under 6GB.
Latest version:
4.0.0
Required dependencies:
accelerate
|
huggingface-hub
|
safetensors
|
scipy
|
sentencepiece
|
torch
|
tqdm
|
transformers
Downloads last day:
378
Downloads last week:
2,445
Downloads last month:
5,466
7-day smoothing