NVIDIA/Megatron-LM
Ongoing research training transformer models at scale
16443 stars
Python
2 awesome lists
facebookresearch/metaseq
Repo for external large-scale work
- Stars
- 6,549
- Forks
- 718
- Commits
- —
- Language
- Python
- Awesome lists
- 1
Similar repositories
NVIDIA/Model-Optimizer
A unified library of SOTA model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.
2759 stars
Python
1 awesome list
facebookresearch/seamless_communication
Foundational Models for State-of-the-Art Speech and Text Translation
11779 stars
Jupyter Notebook
1 awesome list
kvcache-ai/ktransformers
A Flexible Framework for Experiencing Heterogeneous LLM Inference/Fine-tune Optimizations
17195 stars
Python
1 awesome list
OpenNMT/CTranslate2
Fast inference engine for Transformer models
4494 stars
C++
1 awesome list
facebookresearch/fairseq2
FAIR Sequence Modeling Toolkit 2
1135 stars
Python
1 awesome list
Tracked growth
1 capture since 2026-05-27
Latest capture 2026-05-27 12:44
Stars history
Total starsCommits history
Default branch commitsMetadata
- Created: 2022-05-02
- First commit: —
- Last pushed: 2024-04-27
- Archived: yes
- Stack detected: —
- License: MIT
AI development signals
No AI development config files detected.