deepreinforce-ai / CUDA-L1Links
CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning
☆277Updated last month
Alternatives and similar repositories for CUDA-L1
Users that are interested in CUDA-L1 are comparing it to the libraries listed below
Sorting:
- QeRL enables RL for 32B LLMs on a single H100 GPU.☆470Updated last month
- Load compute kernels from the Hub☆354Updated last week
- Simple & Scalable Pretraining for Neural Architecture Research☆305Updated 3 weeks ago
- A collection of tricks and tools to speed up transformer models☆194Updated 2 weeks ago
- FlexAttention based, minimal vllm-style inference engine for fast Gemma 2 inference.☆327Updated last month
- Efficient LLM Inference over Long Sequences☆394Updated 6 months ago
- 👷 Build compute kernels☆196Updated last week
- The Automated LLM Speedrunning Benchmark measures how well LLM agents can reproduce previous innovations and discover new ones in languag…☆122Updated 2 months ago
- Official PyTorch implementation for Hogwild! Inference: Parallel LLM Generation with a Concurrent Attention Cache☆136Updated 4 months ago
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆131Updated last year
- Pytorch Distributed native training library for LLMs/VLMs with OOTB Hugging Face support☆214Updated this week
- Quantized LLM training in pure CUDA/C++.☆224Updated last week
- ☆260Updated 6 months ago
- ☆113Updated last month
- Repo for "LoLCATs: On Low-Rank Linearizing of Large Language Models"☆249Updated 11 months ago
- FlashRNN - Fast RNN Kernels with I/O Awareness☆173Updated 2 months ago
- Work in progress.☆76Updated last month
- The evaluation framework for training-free sparse attention in LLMs☆108Updated 2 months ago
- Training-free Post-training Efficient Sub-quadratic Complexity Attention. Implemented with OpenAI Triton.☆148Updated last month
- Developer Asset Hub for NVIDIA Nemotron — A one-stop resource for training recipes, usage cookbooks, and full end-to-end reference exampl…☆246Updated last week
- Samples of good AI generated CUDA kernels☆95Updated 7 months ago
- Multi-Turn RL Training System with AgentTrainer for Language Model Game Reinforcement Learning☆55Updated last week
- Dion optimizer algorithm☆409Updated last week
- PyTorch implementation of models from the Zamba2 series.☆186Updated 11 months ago
- Official implementation for Training LLMs with MXFP4☆116Updated 8 months ago
- OpenTinker is an RL-as-a-Service infrastructure for foundation models☆424Updated this week
- GRadient-INformed MoE☆265Updated last year
- A safetensors extension to efficiently store sparse quantized tensors on disk☆225Updated last week
- ☆159Updated 6 months ago
- NanoGPT-speedrunning for the poor T4 enjoyers☆73Updated 8 months ago