microsoft / ArchScaleLinks
Simple & Scalable Pretraining for Neural Architecture Research
☆307Updated 2 months ago
Alternatives and similar repositories for ArchScale
Users that are interested in ArchScale are comparing it to the libraries listed below
Sorting:
- Storing long contexts in tiny caches with self-study☆233Updated 2 months ago
- EvaByte: Efficient Byte-level Language Models at Scale☆115Updated 9 months ago
- OpenCoconut implements a latent reasoning paradigm where we generate thoughts before decoding.☆175Updated last year
- Memory optimized Mixture of Experts☆72Updated 6 months ago
- Accelerating your LLM training to full speed! Made with ❤️ by ServiceNow Research☆287Updated this week
- Open-source release accompanying Gao et al. 2025☆501Updated last month
- The Automated LLM Speedrunning Benchmark measures how well LLM agents can reproduce previous innovations and discover new ones in languag…☆128Updated 3 months ago
- Official PyTorch implementation for Hogwild! Inference: Parallel LLM Generation with a Concurrent Attention Cache☆140Updated 5 months ago
- code for training & evaluating Contextual Document Embedding models☆202Updated 8 months ago
- PyTorch implementation of models from the Zamba2 series.☆186Updated last year
- Memory layers use a trainable key-value lookup mechanism to add extra parameters to a model without increasing FLOPs. Conceptually, spars…☆371Updated last year
- Repo for "LoLCATs: On Low-Rank Linearizing of Large Language Models"☆252Updated last year
- an open source reproduction of NVIDIA's nGPT (Normalized Transformer with Representation Learning on the Hypersphere)☆110Updated 11 months ago
- FlexAttention based, minimal vllm-style inference engine for fast Gemma 2 inference.☆334Updated 3 months ago
- MoE training for Me and You and maybe other people☆335Updated last month
- Matrix (Multi-Agent daTa geneRation Infra and eXperimentation framework) is a versatile engine for multi-agent conversational data genera…☆261Updated this week
- PyTorch-native post-training at scale☆605Updated last week
- rl from zero pretrain, can it be done? yes.☆286Updated 4 months ago
- All information and news with respect to Falcon-H1 series☆108Updated 3 months ago
- A scalable asynchronous reinforcement learning implementation with in-flight weight updates.☆361Updated this week
- Reverse Engineering Gemma 3n: Google's New Edge-Optimized Language Model☆262Updated 8 months ago
- This repo contains the source code for the paper "Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning"☆288Updated 2 months ago
- [ICLR 2026] Official PyTorch Implementation of RLP: Reinforcement as a Pretraining Objective☆231Updated last week
- 👷 Build compute kernels☆214Updated last week
- Lightly-reviewed collection of community environments☆210Updated last week
- GRadient-INformed MoE☆264Updated last year
- Code for "LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding", ACL 2024☆356Updated 2 weeks ago
- ☆137Updated 10 months ago
- An extension of the nanoGPT repository for training small MOE models.☆233Updated 10 months ago
- [ACL 2024] Do Large Language Models Latently Perform Multi-Hop Reasoning?☆88Updated 10 months ago