AlexCheema/talos-vs-macbook
microGPT benchmarks: a single M4 Max MacBook Pro P-core in C runs Karpathy's 4192-parameter transformer at ~71x the throughput of TALOS-V2's FPGA implementation.
microGPT benchmarks: a single M4 Max MacBook Pro P-core in C runs Karpathy's 4192-parameter transformer at ~71x the throughput of TALOS-V2's FPGA implementation.
Script is an experimental framework for writing performant trustless trading strategies.
Write Succinct Proofs for Graphs easily
A high-throughput and memory-efficient inference and serving engine for LLMs
vLLM Quantization plugin for GGUF
A trustless wallet scoring protocol
Screenshots for PR comments
MLX: An array framework for Apple silicon
Run LLMs with MLX
Ethereum TypeScript Client Library - for perfect types and tiny builds.
A curated list of awesome projects and papers for distributed training or inference
You like pytorch? You like micrograd? You love tinygrad! ❤️
Examples in the MLX framework
Inference Llama 2 in one file of pure C
How do LLMs change over time?
Veritas is a library for verifying inference of large ML models on Mina.
🦄 🦄 🦄 🦄 Closing the CEX-DEX UX gap, one Uniswap v4 hook at a time
EduDAO Student Hackathon 2023. This is our submission. Team from Oxford University.
MetaMask snap supercharging the Web3 gaming experience with automatic secure transaction signing
Tutorials for creating and using ONNX models
Core smart contracts of Snapshot X for EVM
zkSNARK implementation in JavaScript & WASM
Use Metamask Snaps to automate transaction sending in Metamask
Ethereum peer-to-peer networking specifications
🌐 Human-friendly and powerful HTTP request library for Node.js