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.

baidu-baige/LoongForge

View on GitHub ↗

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.

LoongForge ranks 55th of 370 in Training Toolkits, actively maintained, last commit yesterday.

Star trajectory

22 days tracked
2026-10-31884 stars today2026-11-22
Momentum

- in 30d, below the Training Toolkits median of +0.3%.

Maintenance

Last commit yesterday, fresher than the median of 199.5d.

How it compares in Training Toolkits

LoongForge
884
stars
- 30d
commit yesterday
Training Toolkits median
320
stars
+0.3% 30d
commit ~199.5d ago
Leader · TTS
46.3k
stars
+0.3% 30d
commit 827 days ago

Ranked 55th of 370 by stars.

Ranked by shared tags, then keywords, then stars
  1. Halo is an open-source framework built by White Circle for training large language and multimodal models

    +34 (+4.1%) in 30d · Committed today · Python · Other license

    Shares Pretraining & Training, Training Toolkits, Fine-Tuning (LoRA / PEFT)

  2. Autonomous Qwen3-VL training-code research on the official DocVQA benchmark. main: NVIDIA multi-GPU, mlx…

    +0 (+0.0%) in 30d · Committed 22d ago · Python · MIT

    Shares Pretraining & Training, Training Toolkits, Fine-Tuning (LoRA / PEFT)

  3. An LLM training framework built from the ground up, featuring a custom BumbleBee architecture and end-to-end…

    +0 (+0.0%) in 30d · Committed 90d ago · Python · Apache-2.0

    Shares Pretraining & Training, Training Toolkits, Fine-Tuning (LoRA / PEFT)

  4. Fine-tune an 8B LLM on a 4 GB laptop GPU, check it with real quality gates, and serve it (pip install…

    +0 in 30d · Committed 3d ago · Python · Apache-2.0

    Shares Fine-Tuning (LoRA / PEFT), Pretraining & Training, Training Toolkits

  5. From-scratch PyTorch replications of classic and SOTA AI/ML papers: transformers, attention, GANs, diffusion…

    Committed 17d ago · Jupyter Notebook · Apache-2.0

    Shares Pretraining & Training, Training Toolkits, Fine-Tuning (LoRA / PEFT)

  6. LLM-And-More is a professional, plug-and-play, llm trainer and application builder that guides you through the…

    +6 (+1.6%) in 30d · Committed 870d ago · Go

    Shares Fine-Tuning (LoRA / PEFT), Pretraining & Training, Training Toolkits

  7. A Mac-local specialist-model factory for data, post-training, evaluation, packaging, and evidence-backed…

    +0 (+0.0%) in 30d · Committed yesterday · Swift · MIT

    Shares Training Toolkits, Pretraining & Training, Fine-Tuning (LoRA / PEFT)

  8. A simple, performant, and scalable Jax LLM!

    +25 (+1.0%) in 30d · Committed today · Python · Apache-2.0

    Shares Pretraining & Training, Fine-Tuning (LoRA / PEFT), Training Toolkits

  9. An open-source solution for full parameter fine-tuning of DeepSeek-V3/R1 671B, including complete code and…

    +0 (+0.0%) in 30d · Committed 619d ago · Python · Apache-2.0

    Shares Pretraining & Training, Training Toolkits, Fine-Tuning (LoRA / PEFT)

  10. Easy and lightning fast training of 🤗 Transformers on Habana Gaudi processor (HPU)

    +0 (+0.0%) in 30d · Committed 6d ago · Python · Apache-2.0

    Shares Pretraining & Training, Training Toolkits, Fine-Tuning (LoRA / PEFT)

Language
Python
License
Apache License 2.0
Contributors
39
Forks
70
Watchers
4
Open issues
20
Releases
4
Created
2026-04-23
Stars this week
+38

You've seen where LoongForge stands today.

Get the moves before everyone else: the 5 Training Toolkits repos climbing fastest, every Tuesday.

Free weekly digest. Double opt-in, unsubscribe anytime.

Maintain LoongForge?

Claim this page to add your own call-to-action.

Claim this page