Tag
LiteLLM has migrated from Python to Rust, achieving massive performance improvements: request overhead reduced by 150x to 0.05ms, throughput increased by 15x, memory usage reduced by 11x to 32MB.
The author open-sources their Rust-based browser automation tool with infrastructure and deployment features, originally a business, now free for AI agent development.
Oak is an open-source version control system designed for AI agents, offering branch-per-session workflows, content-addressed lazy mounts, and faster performance than git. It is in public beta and available as a CLI tool and reusable Rust library.
Kyde is a fast native Git client and code editor built in Rust using Zed's gpui framework, featuring GPU rendering, side-by-side diffs, tree-sitter syntax highlighting, and a hand-tuned dark theme.
Introducing rmux, a modern terminal multiplexer written in Rust, designed for AI Agent and automation scenarios. It coordinates multiple Agents running simultaneously, is compatible with over 90 tmux commands, and offers a Python SDK and browser-based session sharing.
agent-browser is a CLI tool for browser automation designed for AI agents, using compact text output and ref-based element selection to minimize token usage. The post also highlights three other tools—portless, emulate, and ai-cli—for improving agent loop efficiency.
An experiment on using Rust's type system as an AI coding specification, distilling practices from Epic Lora.
A Rust developer profiles and optimizes the incremental rebuild time of SQLx tests, identifying bottlenecks like debuginfo generation and proc macro overhead, and proposes improvements to speed up test compilation.
Introduces oh-my-pi, a terminal-based AI coding assistant forked from Pi. With its Hashline editing system, it achieves high-precision code modifications, reducing token consumption by 61%. It comes with 32 built-in tools, supports over 40 languages and more than 40 LLM providers, and can import configurations directly from major tools, making migration costs negligible.
A high-performance instant messaging backend based on Rust + Actix-web + QUIC, supporting P2P hole punching, group chat broadcasting, multiple storage backends, can be deployed as a single process or separately.
Noema Atlas is a free and open-source peer-to-peer desktop app for decentralized distribution of LLM model weights, using content-addressed verification and Iroh for direct machine-to-machine transfers, with Hugging Face as a fallback.
Technical deep dive into Lume, a Rust hybrid search engine that combines BM25, dense vectors, and entity graphs for auditable, local-first retrieval.
cargo-geiger is a Rust cargo plugin that lists statistics about unsafe code usage in a crate and its dependencies, providing input for auditing.
A Lisp interpreter embedded in Rust's trait system, allowing recursive functions, closures, and continuation passing style at compile time.
Rust's SIMD abstractions now allow safe usage without unsafe code by leveraging CPU feature tokens introduced in Rust 1.87, enabling concise and portable vector operations.
Bevy 0.19, a major update to the Rust game engine, introduces next-generation scenes (BSN format), faster rendering with more GPU work, contact shadows, new UI widgets, text input support, post-processing effects, and more.
A Chinese developer created Pake, a tool using Rust and Tauri to build ultra-lightweight desktop apps from webpages, dramatically reducing memory usage compared to Electron-based apps. The open-source project has gained significant traction with over 51,000 GitHub stars.
A developer shares their weekend project of building a low-level infix language that compiles to WebAssembly, and offers a personal ranking of AI coding tools from contextual autocomplete to frontier models.
cuTile Rust introduces a tile-based programming model that leverages Rust's ownership to guarantee memory safety and data-race freedom for GPU kernels, and the Grout inference engine built on it achieves competitive throughput with vLLM/SGLang for Qwen3 models.
mistral.rs v0.8.10 adds /v1/skills support, enabling agent skills execution, along with other improvements.