PrimeIntellect-ai / pcclLinks
PCCL (Prime Collective Communications Library) implements fault tolerant collective communications over IP
☆141Updated 4 months ago
Alternatives and similar repositories for pccl
Users that are interested in pccl are comparing it to the libraries listed below
Sorting:
- SIMD quantization kernels☆94Updated 4 months ago
- Write a fast kernel and run it on Discord. See how you compare against the best!☆68Updated this week
- Ship correct and fast LLM kernels to PyTorch☆139Updated 2 weeks ago
- 👷 Build compute kernels☆214Updated last week
- MoE training for Me and You and maybe other people☆331Updated last month
- NSA Triton Kernels written with GPT5 and Opus 4.1☆70Updated 5 months ago
- train with kittens!☆63Updated last year
- Modded vLLM to run pipeline parallelism over public networks☆40Updated 8 months ago
- PTX-Tutorial Written Purely By AIs (Deep Research of Openai and Claude 3.7)☆66Updated 10 months ago
- DeMo: Decoupled Momentum Optimization☆198Updated last year
- Quantized LLM training in pure CUDA/C++.☆235Updated 2 weeks ago
- Storing long contexts in tiny caches with self-study☆233Updated last month
- ☆219Updated last year
- Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters☆131Updated last year
- Hand-Rolled GPU communications library☆81Updated 2 months ago
- ring-attention experiments☆165Updated last year
- The Automated LLM Speedrunning Benchmark measures how well LLM agents can reproduce previous innovations and discover new ones in languag…☆127Updated 3 months ago
- Official CLI and Python SDK for Prime Intellect - access GPU compute, remote sandboxes, RL environments, and distributed training infrast…☆148Updated this week
- Docker image NVIDIA GH200 machines - optimized for vllm serving and hf trainer finetuning☆53Updated 11 months ago
- an open source reproduction of NVIDIA's nGPT (Normalized Transformer with Representation Learning on the Hypersphere)☆110Updated 10 months ago
- ☆47Updated 2 years ago
- Memory optimized Mixture of Experts☆72Updated 6 months ago
- Official code for "SWARM Parallelism: Training Large Models Can Be Surprisingly Communication-Efficient"☆149Updated 2 years ago
- ☆237Updated 3 weeks ago
- Long context evaluation for large language models☆225Updated 11 months ago
- ☆22Updated last year
- Samples of good AI generated CUDA kernels☆99Updated 8 months ago
- Small scale distributed training of sequential deep learning models, built on Numpy and MPI.☆155Updated 2 years ago
- NanoGPT (124M) quality in 2.67B tokens☆28Updated 4 months ago
- train entropix like a champ!☆20Updated last year