facebookresearch / linguaLinks
Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.
☆4,750Updated 5 months ago
Alternatives and similar repositories for lingua
Users that are interested in lingua are comparing it to the libraries listed below
Sorting:
- NanoGPT (124M) in 3 minutes☆4,035Updated last week
- A PyTorch native platform for training generative AI models☆4,892Updated this week
- PyTorch native post-training library☆5,639Updated this week
- Minimalistic 4D-parallelism distributed training framework for education purpose☆1,935Updated 4 months ago
- Code for BLT research paper☆2,018Updated last month
- Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.☆6,172Updated 4 months ago
- The simplest, fastest repository for training/finetuning small-sized VLMs.☆4,468Updated 2 months ago
- Sky-T1: Train your own O1 preview model within $450☆3,366Updated 5 months ago
- Minimalistic large language model 3D-parallelism training☆2,396Updated 3 weeks ago
- Entropy Based Sampling and Parallel CoT Decoding☆3,434Updated last year
- Implementing DeepSeek R1's GRPO algorithm from scratch☆1,724Updated 8 months ago
- Large Concept Models: Language modeling in a sentence representation space☆2,317Updated 11 months ago
- Everything about the SmolLM and SmolVLM family of models☆3,516Updated last month
- Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.☆2,795Updated last week
- Repository for Meta Chameleon, a mixed-modal early-fusion foundation model from FAIR.☆2,073Updated last year
- Schedule-Free Optimization in PyTorch☆2,244Updated 7 months ago
- Our library for RL environments + evals☆3,674Updated this week
- AllenAI's post-training codebase☆3,488Updated this week
- DataComp for Language Models☆1,403Updated 3 months ago
- [ICLR 2025] Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling☆936Updated last month
- Modeling, training, eval, and inference code for OLMo☆6,263Updated last month
- Distilabel is a framework for synthetic data and AI feedback for engineers who need fast, reliable and scalable pipelines based on verifi…☆3,008Updated last week
- Tools for merging pretrained large language models.☆6,647Updated 2 weeks ago
- Bringing BERT into modernity via both architecture changes and scaling☆1,598Updated 6 months ago
- Implementation for MatMul-free LM.☆3,043Updated last month
- Training Large Language Model to Reason in a Continuous Latent Space☆1,419Updated 4 months ago
- Democratizing Reinforcement Learning for LLMs☆4,917Updated this week
- Lighteval is your all-in-one toolkit for evaluating LLMs across multiple backends☆2,226Updated 2 weeks ago
- Efficient Triton Kernels for LLM Training☆5,991Updated last week
- ☆3,054Updated last month