facebookresearch / chai
CHAI is a library for dynamic pruning of attention heads for efficient LLM inference.
☆12Updated 3 months ago
Alternatives and similar repositories for chai:
Users that are interested in chai are comparing it to the libraries listed below
- ☆31Updated last year
- Official repo of dataset-decomposition paper [NeurIPS 2024]☆13Updated 2 months ago
- ☆33Updated 6 months ago
- GoldFinch and other hybrid transformer components☆10Updated 2 weeks ago
- Implementation of the LDP module block in PyTorch and Zeta from the paper: "MobileVLM: A Fast, Strong and Open Vision Language Assistant …☆16Updated last year
- JAX Scalify: end-to-end scaled arithmetics☆16Updated 5 months ago
- An open source replication of the stawberry method that leverages Monte Carlo Search with PPO and or DPO☆28Updated 3 weeks ago
- Train a SmolLM-style llm on fineweb-edu in JAX/Flax with an assortment of optimizers.☆17Updated 2 weeks ago
- Exploring an idea where one forgets about efficiency and carries out attention across each edge of the nodes (tokens)☆47Updated last week
- Training hybrid models for dummies.☆20Updated 2 months ago
- Implementation of Diffusion Transformers and Rectified Flow in Jax☆21Updated 8 months ago
- Triton kernels for Flux☆20Updated 3 months ago
- GoldFinch and other hybrid transformer components☆45Updated 8 months ago
- Implementation of Google's USM speech model in Pytorch☆30Updated 2 months ago
- Code associated with the paper: CTC-DRO: Robust Optimization for Reducing Language Disparities in Speech Recognition.☆14Updated 3 weeks ago
- QuIP quantization☆52Updated last year
- This library supports evaluating disparities in generated image quality, diversity, and consistency between geographic regions.☆20Updated 9 months ago
- [Oral; Neurips OPT2024 ] μLO: Compute-Efficient Meta-Generalization of Learned Optimizers☆12Updated 2 weeks ago
- implementation of https://arxiv.org/pdf/2312.09299☆20Updated 8 months ago
- Here we will test various linear attention designs.☆60Updated 11 months ago
- Official Code Implementation for 'A Simple Early Exiting Framework for Accelerated Sampling in Diffusion Models'☆16Updated 8 months ago
- A JAX library for building lattice-based speech transducer models☆43Updated 3 months ago
- ☆10Updated 10 months ago
- Engineering the state of RNN language models (Mamba, RWKV, etc.)☆32Updated 10 months ago
- The official code of "Building on Efficient Foundations: Effectively Training LLMs with Structured Feedforward Layers"☆19Updated 8 months ago
- research impl of Native Sparse Attention (2502.11089)☆53Updated last month
- Zeta implementation of a reusable and plug in and play feedforward from the paper "Exponentially Faster Language Modeling"☆15Updated 4 months ago
- Exploration into the proposed "Self Reasoning Tokens" by Felipe Bonetto☆55Updated 10 months ago
- Code for the note "NF4 Isn't Information Theoretically Optimal (and that's Good)☆18Updated last year
- Implementation of Token Shift GPT - An autoregressive model that solely relies on shifting the sequence space for mixing☆48Updated 3 years ago