Stars
Token-efficient schema definition for getting structured output from LLMs.
🎒 Token-Oriented Object Notation (TOON) – Compact, human-readable, schema-aware JSON for LLM prompts. Spec, benchmarks, TypeScript SDK.
Accelerating MoE with IO and Tile-aware Optimizations
MoE training for Me and You and maybe other people
Mcrouter is a memcached protocol router for scaling memcached deployments.
Versatile typeface for code, from code.
Benchmark that evaluates LLMs using 759 NYT Connections puzzles extended with extra trick words
Clojure library for fake data generation, port of ruby faker
Advanced quantization toolkit for LLMs and VLMs. Support for WOQ, MXFP4, NVFP4, GGUF, Adaptive Schemes and seamless integration with Transformers, vLLM, SGLang, and llm-compressor
A Clojurey wrapper around the Lanterna terminal output library.
Sourcecodes of our Evolutionary Reasoning Optimization project.
DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
FlashMLA: Efficient Multi-head Latent Attention Kernels
This repository allows reproduction of Poetiq's record-breaking submission to the ARC-AGI-1 and ARC-AGI-2 benchmarks.
SWE-bench: Can Language Models Resolve Real-world Github Issues?
The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-bench verified!
SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024]
Implements ristretto255, a fast prime-order group.
High-performance scale-up self-hosted simple storage service (hs5)
SONIC: Spectral Optimization of Noise for Inpainting with Consistency
A library for reading and writing the DWARF debugging format