GitHub Trending repo Repositories Jul 5 3 views

avifenesh/bw24: From-scratch Rust+CUDA LLM inference engine for RTX 5090 (sm_120a) — NVFP4, MoE, MTP speculative decoding, tuned against measured hardware limits

by avifenesh

From-scratch Rust+CUDA LLM inference engine for RTX 5090 (sm_120a) — NVFP4, MoE, MTP speculative decoding, tuned against measured hardware limits

Read Original

AI Hardware LLM NVIDIA

Metadata

Stars: 91
Forks: 1
Language: Rust
Watchers: 91
License: MIT

GitHub Trending repo 11h ago

alphaparkinc/genpark-multimodal-generative-art-prompt-synthesis-engine-skill: Multimodal generative art prompt synthesis & diffusion renderer (SeaArt style)

Multimodal generative art prompt synthesis & diffusion renderer (SeaArt style)

GitHub Trending repo 20h ago

Greninja9257/LabLLM: A native macOS lab for teaching tiny language models to think — build the architecture, train the weights, and watch a small LLM emerge from scratch, locally on Apple Silicon with custom data, tokenizers, checkpoints, and MLX acceleration.

A native macOS lab for teaching tiny language models to think — build the architecture, train the weights, and watch a small LLM emerge from scratch, locally on Apple Silicon with ...

GitHub Trending repo 1d ago

cneuralnetwork/oplogs: Local-first, open-source experiment tracking for machine learning and agents

Local-first, open-source experiment tracking for machine learning and agents

avifenesh/bw24: From-scratch Rust+CUDA LLM inference engine for RTX 5090 (sm_120a) — NVFP4, MoE, MTP speculative decoding, tuned against measured hardware limits

Metadata

Related

alphaparkinc/genpark-multimodal-generative-art-prompt-synthesis-engine-skill: Multimodal generative art prompt synthesis & diffusion renderer (SeaArt style)

Greninja9257/LabLLM: A native macOS lab for teaching tiny language models to think — build the architecture, train the weights, and watch a small LLM emerge from scratch, locally on Apple Silicon with custom data, tokenizers, checkpoints, and MLX acceleration.

cneuralnetwork/oplogs: Local-first, open-source experiment tracking for machine learning and agents