AIChat is an all-in-one LLM command-line tool, written in Rust and licensed under Apache-2.0, that gives developers a shell assistant, an interactive chat REPL, retrieval-augmented generation, and AI tools and agents through a single terminal interface to more than twenty language model providers.

What it is

AIChat is a command-line application in the Rust ecosystem, distributed through crates.io and the usual system package managers, that unifies access to over twenty language model providers behind one command. It connects to OpenAI, Claude, Gemini (Google AI Studio), Ollama, Groq, Azure-OpenAI, VertexAI, Bedrock, Github Models, Mistral, Deepseek, AI21, XAI Grok, Cohere, Perplexity, Cloudflare, OpenRouter, Ernie, Qianwen, Moonshot, ZhipuAI, MiniMax, Deepinfra, and VoyageAI, and it can also address any provider that exposes an OpenAI-compatible API. The same binary covers one-shot command-line prompts, an interactive REPL, and, per its repository topics, a web UI, which means the choice of model becomes a configuration detail rather than a reason to learn a different tool.

The concrete problem it solves is fragmentation. Anyone working with more than one model vendor otherwise maintains separate credentials, separate client programs, and separate invocation habits for each provider, plus ad-hoc shell scripts for anything that needs to pipe in local files or command output. AIChat replaces that collection of provider-specific clients and throwaway scripts with one aichat command whose flags and input forms stay the same regardless of which model is answering, and it lives in the AI Interaction and Interfaces category of this registry.

Key capabilities

• Multi-provider access through a unified interface covering more than twenty providers, including OpenAI, Claude, Gemini, Ollama, Groq, Azure-OpenAI, VertexAI, Bedrock, and any OpenAI-compatible API endpoint.
• CMD mode for one-shot use, invoked as aichat hello or by piping input such as cat data.txt | aichat.
• REPL mode, an interactive chat loop with tab autocompletion, multi-line input, history search, configurable keybindings, and custom REPL prompts.
• Shell Assistant, which turns a task described in natural language into a shell command and adjusts to the host operating system and shell environment.
• Multi-form input through the -f flag and the REPL .file command: local files such as aichat -f image.png -f data.txt, whole directories as aichat -f dir/, remote URLs as aichat -f https://example.com, external command output such as git diff, the last reply via .file %%, and combined inputs such as aichat -f dir/ -f data.txt explain.
• Roles and sessions, where a role bundles a prompt with model configuration and a session preserves context across a conversation.
• Macros, RAG for pulling external documents into a conversation, and function calling, with supporting functions hosted in the separate llm-functions repository.

Who uses it and how

• Terminal-centric developers who want natural-language task descriptions converted into commands matched to their own operating system and shell, without leaving the prompt.
• Engineers who switch between hosted and local models, using OpenAI, Claude, or Gemini keys for heavy work and Ollama for local inference under the same command.
• Users who automate: CMD mode composes inside shell pipelines, and macros collapse repeated sequences of REPL commands into one invocation for recurring tasks.
• Practitioners doing document-grounded work, who point RAG at directories or files rather than pasting content by hand.
• Android users on Termux, who install through pkg install aichat, alongside Homebrew, Pacman, and Scoop users on desktop systems.

Getting started

Install with cargo install aichat, brew install aichat, pacman -S aichat, scoop install aichat, or pkg install aichat on Termux; pre-built binaries for macOS, Linux, and Windows are also available from GitHub Releases and need only to be placed on the PATH. Configuration is then a matter of adding provider credentials or pointing at a local Ollama instance.

How it compares

This registry entry names no comparable or competing products, so AIChat stands alone here rather than being contrasted with a list of paid alternatives. A reader evaluating it against another tool should do so outside this page.

When to use it — and when not to

Choose AIChat when the goal is one terminal workflow across many providers and the operator is comfortable supplying their own API credentials or running a local model server. It is the wrong fit for anyone seeking a managed, hosted service with provider billing handled for them, and it should be treated with care where function calling matters, since the function definitions live in a separate repository rather than inside this project. The entry carries a substantial open issue count of 104, and the README excerpt shown in this registry is truncated, so a prospective adopter should read the upstream documentation in full before standardizing on it.

$ aichat --serve
Chat Completions API: http://127.0.0.1:8000/v1/chat/completions
Embeddings API:       http://127.0.0.1:8000/v1/embeddings
Rerank API:           http://127.0.0.1:8000/v1/rerank
LLM Playground:       http://127.0.0.1:8000/playground
LLM Arena:            http://127.0.0.1:8000/arena?num=2

curl -X POST -H "Content-Type: application/json" -d '{
  "model":"claude:claude-3-5-sonnet-20240620",
  "messages":[{"role":"user","content":"hello"}], 
  "stream":true
}' http://127.0.0.1:8000/v1/chat/completions