lyogavin/airllm
AirLLM 2.8T LLM inference with single 4GB GPU
Older snapshot: check the original source for current information.
From the publisher
README & documentation
Source previewRead the project’s overview, installation instructions and usage examples. The original README is the source of truth.
Quickstart | Configurations | MacOS | Example notebooks | FAQ AirLLM dramatically reduces inference memory usage, letting 70B large language models run on a single 4GB GPU card — without quantization, distillation, or pruning. You can even run Kimi K3 (2.8T) — the largest open-source model released to date — on under 4GB, Qwen3.8-Flash-Next (125B) on 6GB, and DeepSeek-V3 (671B) on 12GB. We now also support training huge models on small VRAM: Qwen3.8-Flash-Next (125B) under 6GB. [](https://pepy.tech/project/airllm) [](https://github.com/LianjiaTech/BELLE/blob/main/LICENSE) [](https://static.aicompose.cn/static/wecombarcode.png?t=1671918938) [](https://discord.gg/2xffU5sn) [](https://medium.com/@lyo.gavin) [](https://gavinliblog.com) [](https://patreon.com/gavinli) [](https://github.com/sponsors/lyogavin) [2026/09] Training support: stream frozen weights one layer at a time and keep adapters on the GPU. Qwen3.8-Flash-Next (125B) trains under 6GB (RTX 3060 Ti); Qwen3.8-27B trains in 2GB at seq 512. See Training. [2026/08] Qwen3.8-Flash-Next support: Qwen's 125B MoE flagship (Qwen4ExpForConditionalGeneration) with a 51B n-gram embedding runs in 5.95GB of VRAM, measured end to end on one RTX 4090. The n-gram table is file-mapped on…
Read the full README ↗ · Preview checked 2026-10-31T04:48:09.133Z
Inside the original README — Document outline
- AI Agents Recommendation:
- Updates
- Star History
- Table of Contents
- Quickstart
- 1. Install package
- 2. Inference
- Model Compression - 3x Inference Speed Up!
- How to enable model compression speed up:
- What are the differences between model compression and quantization?
- Configurations
- MacOS
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Links from the README
References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.
Documentation belongs to its respective authors. Reported project/model license: Apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
What this repository does
AirLLM 2.8T LLM inference with single 4GB GPU
Repository facts
- Owner
- lyogavin
- Primary language
- Jupyter Notebook
- Stars
- 37,977
- Forks
- 3,895
- Open issues + pull requests
- 157
- License
- Apache-2.0
- Archived
- No
- Default branch
- main
- Created
- 2023-06-12T16:28:41.000Z
- Last push
- 2026-10-11T06:19:49.000Z
Topics and intended use
Owner-supplied topics: chinese-llm, chinese-nlp, finetune, generative-ai, instruct-gpt, instruction-set, llama, llm, lora, open-models, open-source, open-source-models, qlora
Review the README for scope, installation, examples and limitations. We do not run repository code or certify it.
Evaluate before installing
Review licensing and dependencies, inspect recent commits and unresolved issues, and test in an isolated environment before production use. Stars and forks alone cannot answer these questions.
Source and freshness
Source: GitHub. Metadata observed 2026-10-13T12:06:01.583Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.