baidu-baige / LoongForge
A high-performance framework for training LLMs, VLMs, diffusion, and embodied models on NVIDIA GPUs and Kunlun XPUs. Supports Pi0.5, GR00T-N1.7, FastWAM, Cosmos3, DreamZero, DeepSeek-V4, GLM-5.3, Kimi-K3 and more.
Why this job: matched by what the repo talks about: llm, gpu, model, performance
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit today
- Stars
- 839
- +17 this week
- Licence
- Apache-2.0
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- v0.2.2
- Released 7 days ago
- Built with
- Python
- 94% of the code
Is it looked after?
- Contributors
- 26
- Last 90 days
- Issues
- 66% closed
- 19 still open
- Pull requests
- 242 merged
- 13 waiting
- Releases
- 4
- 4 in the last year
- Contributing guide (opens in new tab)
- No security policy
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| mingti-org / phyai | 152 | yesterday | MIT | No downloads |
| ViperEkura / AstrAI | 111 | yesterday | Apache-2.0 | Linux |
More: languages, topics and project record
Languages
- Python 94%
- Shell 3%
- Cuda 2%
- Jinja 1%
Topics
- llm
- vla
- vlm
- wan
- pretraining
- sft
- megatron
- lora
- mid-training
- wam
- gpu
- kunlun
Project record
- JobLocal LLMs
- StatusListed
- Commits514
- Forks68
- Watching3
- Stars in 28 days+159
- CreatedApr 23, 2026
- First seen by usNov 7, 2026
- CheckedNov 12, 2026