lyogavin/airllm
Low memory LLM inference tool
AirLLM enables the inference and training of massive large language models on low-memory GPUs by drastically reducing memory usage without quantization or pruning. It allows running very large open-source models on single consumer-grade hardware.
- Stars
- 34,800
- Stars gained in 7 days
- +300
- Contributors
- 10
- Forks
- 3,700
- Days trending in the last 30
- 4 days
- Mark
- Possible hype — 3,480 stars per contributor. An automatic mark on public metrics: above 500 is watch, above 2,000 possible hype (not a judgement of quality).