Eric Buehler

mistral.rs. Fast, flexible LLM inference

candle-vllm. Efficent platform for inference and serving local LLMs including an OpenAI compatible API server.

xlora. X-LoRA: Mixture of LoRA Experts

candle-lora. Low rank adaptation (LoRA) for Candle.

diffusion-rs. Blazingly fast inference of diffusion models.

trc. A faster Arc.

float8. 8-bit floating point types for Rust

safetensors_explorer. CLI utility to inspect and explore .safetensors and .gguf files

candle-sampling. Sampling techniques for Candle.

The-Neuron. An autonomous robot, powered by AI.

Falcon-Programming-Language. Falcon is a powerful, interpreted programming language.

candle_graphs. Graph model execution API for Candle

candle. Minimalist ML framework for Rust

zig_ml. Tensor library for Zig

float4. MXFP4-compatible 4-bit floating point types and block formats for Rust.

constensor. ML framework featuring compile time checks and accelerated by a JIT compiler.

kestrel. Simple and safe.

PerceiverIO-Classifier. A classifier based on PerceiverIO

Kestrel-Programming-Language. The Kestrel Programming Language.

merlin. Make your code fly.

ggml-ffi. Up-to-date Rust FFI bindings to GGML.

CodingGPT. A GPT-like autoregressive, decoder-only model trained on data about various scientific or coding topics.

candle-lora-macro. Macros for candle-lora.

Automatic-Differentiation-Custom. Autodiff from the ground up.

uqff_maker. Automated generation of UQFF models with mistral.rs.

candle-benching. Benchmarking for Candle

Proofling. An automatic proof checker

candle_llm_dataset. A LLM-specific dataset for Candle.

edge-u-cation. Edge(u)cation: Cutting-edge multimodal LLMs on the edge with mistral.rs, using F8Q8

git-persona. A CLI tool to manage and switch between Git user profiles

peft. 🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

flexrc. Fast, flexible, and safe reference counted type.

english-transformer. An autoregressive, decoder-only transformer.

simd_matmul. O(n^2) matmul with SIMD.

once_arc. Initilize and allocate once, deallocate never.

MathGPT. My custom decoder-only, autoregressive, GPT-style model.

candle-cublaslt. Rust

Graph-Aware-Transformers. Graph-Aware Attention for Adaptive Dynamics in Transformers

inkwell. It's a New Kind of Wrapper for Exposing LLVM (Safely)

range-checked. Range checked types leveraging Rust's type system.

pyo3_smd_example. Example of pyo3_special_method_derive

mlx. MLX: An array framework for Apple silicon

merlin_rust. Make your code fly.

dora. DORA (Dataflow-Oriented Robotic Application) is middleware designed to streamline and simplify the creation of AI-based robotic applications. It offers low latency, composable, and distributed dataflow capabilities. Applications are modeled as directed graphs, also referred to as pipelines.

md_check. Check markdown link validity.

reverse-engineering-gemma-3n. Reverse Engineering Gemma 3n: Google's New Edge-Optimized Language Model

candle-layer-norm. Cuda

PyParser. Rust

llguidance. Super-fast Structured Outputs