ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward experts. We released a collection of ModuleFormer-based Language Models (MoLM) ranging in scale from 4 billion to 8 billion parameters.
☆226Sep 18, 2025Updated 5 months ago
Alternatives and similar repositories for ModuleFormer
Users that are interested in ModuleFormer are comparing it to the libraries listed below
Sorting:
- Official repository for the paper "SwitchHead: Accelerating Transformers with Mixture-of-Experts Attention"☆102Sep 30, 2024Updated last year
- ☆415Nov 2, 2023Updated 2 years ago
- Code for the paper "QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models".☆279Nov 3, 2023Updated 2 years ago
- Triton-based implementation of Sparse Mixture of Experts.☆266Oct 3, 2025Updated 4 months ago
- A family of open-sourced Mixture-of-Experts (MoE) Large Language Models☆1,660Mar 8, 2024Updated last year
- ☆274Oct 31, 2023Updated 2 years ago
- some common Huggingface transformers in maximal update parametrization (µP)☆87Mar 14, 2022Updated 3 years ago
- [ICLR 2024] Lemur: Open Foundation Models for Language Agents☆555Oct 28, 2023Updated 2 years ago
- Layer-Condensed KV cache w/ 10 times larger batch size, fewer params and less computation. Dramatic speed up with better task performance…☆156Apr 7, 2025Updated 10 months ago
- Reaching LLaMA2 Performance with 0.1M Dollars☆988Jul 23, 2024Updated last year
- Simplex Random Feature attention, in PyTorch☆76Oct 10, 2023Updated 2 years ago
- GRadient-INformed MoE☆264Sep 25, 2024Updated last year
- ☆20May 30, 2024Updated last year
- [ICLR 2025 & COLM 2025] Official PyTorch implementation of the Forgetting Transformer and Adaptive Computation Pruning☆140Dec 19, 2025Updated 2 months ago
- Repo for "Monarch Mixer: A Simple Sub-Quadratic GEMM-Based Architecture"☆562Dec 28, 2024Updated last year
- ☆33Oct 4, 2024Updated last year
- ☆14Feb 1, 2024Updated 2 years ago
- Repo for "LoLCATs: On Low-Rank Linearizing of Large Language Models"☆250Jan 31, 2025Updated last year
- ☆91Aug 18, 2024Updated last year
- [ICLR 2024] Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning☆641Mar 4, 2024Updated last year
- Fine-Tuning Pre-trained Transformers into Decaying Fast Weights☆19Oct 9, 2022Updated 3 years ago
- ☆16Dec 9, 2023Updated 2 years ago
- [ICML 2024] CLLMs: Consistency Large Language Models☆412Nov 16, 2024Updated last year
- YaRN: Efficient Context Window Extension of Large Language Models☆1,673Apr 17, 2024Updated last year
- [NAACL 2025] A Closer Look into Mixture-of-Experts in Large Language Models☆60Feb 7, 2025Updated last year
- Inference code for Persimmon-8B☆412Sep 9, 2023Updated 2 years ago
- ☆16Apr 10, 2025Updated 10 months ago
- Serving multiple LoRA finetuned LLM as one☆1,145May 8, 2024Updated last year
- train with kittens!☆63Oct 25, 2024Updated last year
- Code repository for the paper "MrT5: Dynamic Token Merging for Efficient Byte-level Language Models."☆54Sep 25, 2025Updated 5 months ago
- Positional Skip-wise Training for Efficient Context Window Extension of LLMs to Extremely Length (ICLR 2024)☆209May 20, 2024Updated last year
- Triton Implementation of HyperAttention Algorithm☆48Dec 11, 2023Updated 2 years ago
- Expanding linear RNN state-transition matrix eigenvalues to include negatives improves state-tracking tasks and language modeling without…☆20Mar 15, 2025Updated 11 months ago
- Official PyTorch implementation of QA-LoRA☆145Mar 13, 2024Updated last year
- Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads☆2,708Jun 25, 2024Updated last year
- DSIR large-scale data selection framework for language model training☆270Apr 7, 2024Updated last year
- [ACL 2023 Findings] Emergent Modularity in Pre-trained Transformers☆26Jun 7, 2023Updated 2 years ago
- PB-LLM: Partially Binarized Large Language Models☆156Nov 20, 2023Updated 2 years ago
- ☆14Nov 20, 2022Updated 3 years ago