bersisse/cast_client — explained in plain English
Analysis updated 2026-05-18
Chat with any OpenAI-compatible AI provider, including Gemini or local Ollama models, from a lightweight desktop app.
Keep multiple saved conversations with an AI assistant without a browser tab consuming hundreds of megabytes of RAM.
Build the app from source on Windows or Linux and point it at a self-hosted or third-party AI endpoint.
| bersisse/cast_client | 132ikl/game | 1lystore/pay-dcp | |
|---|---|---|---|
| Stars | 2 | 2 | 2 |
| Language | Rust | Rust | Rust |
| Last pushed | — | 2020-12-30 | — |
| Maintenance | — | Dormant | — |
| Setup difficulty | moderate | moderate | moderate |
| Complexity | 2/5 | 2/5 | 3/5 |
| Audience | general | general | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires the Rust toolchain to build from source, plus an API key from an OpenAI-compatible provider.
Cast Client is a desktop app for chatting with AI models, built in Rust using the egui interface toolkit instead of a web browser engine. It works with any AI service that offers an OpenAI compatible API, which covers providers like OpenAI and Gemini as well as locally run models through Ollama, so you are not locked into a single AI company's product. The app streams responses as they are generated and renders them as formatted markdown while they arrive, rather than waiting for the full reply. It keeps multiple separate conversations and saves them to disk so you can return to earlier chats. Networking runs on a single background thread rather than a larger pool of worker threads, which the author says keeps the app lightweight. The main reason this project exists is memory usage. The author compared how much RAM different AI chat tools use while idle: a Gemini web page open in a Firefox tab used about 580 megabytes, another coding focused AI tool used about 670 megabytes, while Cast Client used about 110 megabytes on Windows 11. The author built this specifically to see if a native, non browser based chat client could use noticeably less memory, while acknowledging that browser based tools are more full featured and that Cast Client is not trying to replace dedicated coding assistants. To build it yourself you need the Rust toolchain and run a standard cargo build command, though the author notes the memory allocator used may need extra system packages on Linux, and Linux has not been tested as thoroughly as Windows. After building or launching the app, you open its settings and enter a base URL for your chosen AI provider, an API key, and the model name you want to use. The project is open to contributions through pull requests and issues, and the author mentions hoping the project grows over time. The project is released under the GPL 3.0 license.
Cast Client is a lightweight native desktop AI chat app built in Rust that connects to any OpenAI-compatible provider while using far less memory than browser-based alternatives.
Mainly Rust. The stack also includes Rust, egui.
GPL-3.0: you can use and modify the code, but any distributed derivative work must also be open sourced under the same license.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly general.
This repo across BitVibe Labs
double-check against the repo, no cap.