{
 "repo": "lyogavin/airllm",
 "url": "https://github.com/lyogavin/airllm",
 "what": "AirLLM cuts inference memory so very large language models run on a single small GPU without quantization, distillation or pruning, by streaming model layers instead of loading the whole model. The README reports 70B models on a 4GB card, DeepSeek-V3 (671B) on about 12GB and Kimi K3 on under 4GB, all through one AutoModel API, plus optional 4-bit or 8-bit compression for speed and new training support for huge models on small VRAM.",
 "language": "Jupyter Notebook",
 "stars": null,
 "gain": null,
 "created": "",
 "uc_n": 3,
 "title": "Airllm",
 "show": true,
 "what_source": "deduped",
 "use_cases": "https://dedupedai.com/use-cases/lyogavin-airllm/",
 "use_cases_json": "https://dedupedai.com/use-cases/r/lyogavin-airllm.json",
 "json": "https://dedupedai.com/repos/r/lyogavin-airllm.json",
 "slug": "lyogavin-airllm",
 "api": "https://dedupedai.com/api/v1/repos/lyogavin-airllm.json",
 "api_version": "v1"
}