Tác nhân & Đa tác nhân

models

reyamira/models

A TUI and CLI tool for browsing AI models, benchmarks, coding agents, and provider statuses.

★ 492Số sao
⑂ 19Fork
0Vấn đề đang mở
RustNgôn ngữ
MITGiấy phép
Q92Điểm biên tập

Ảnh chụp màn hình dự án

Ảnh chụp màn hình của models Ảnh chụp màn hình của models

Tổng quan

Models is a Rust-based terminal user interface (TUI) and command-line interface (CLI) application that aggregates data from multiple sources to help users browse over 4,000 AI models across 85+ providers, compare roughly 1,000 benchmark entries, track 12+ coding agents, and monitor the live status of 22 AI providers. It features a three-column layout for model browsing, head-to-head benchmark comparisons with scatter plots and radar charts, in-app agent management and updates, and live provider health dashboards.

Tính năng chính

  • Browse 4,000+ models across 85+ providers with filtering and sorting.
  • Compare 1,000 benchmark entries across 4 data sources with scatter plots and radar charts.
  • Track 12+ AI coding agents with automatic version detection and in-app updates.
  • Monitor live health status for 22 AI providers across 7 status page platforms.
  • Interactive TUI with mouse support, context-aware help, and field glossary.
  • Full CLI parity for scripting and automation, including JSON output.
  • In-app addition and updating of custom coding agents.

Yêu cầu, cài đặt và bắt đầu nhanh

Homebrew (macOS/Linux): `brew install models`. Cargo (any platform): `cargo install modelsdev`. Nix (flake): `nix run github:reyamira/models`. Scoop (Windows): `scoop install extras/models`. AUR (Arch Linux): `paru -S models-bin`. Pre-built binaries, .deb, and .rpm packages are available on GitHub Releases.

Cách sử dụng

Launch the TUI by running `models`. Navigate with arrow keys, switch tabs using `[` or `]`, search with `/`, and press `?` for help. Mouse interaction is supported for selecting rows, focusing panels, and scrolling. CLI usage includes commands like `models list`, `models show`, `models search`, `models providers`, `models benchmarks list`, `models benchmarks show`, `agents status`, and `models status list`.

Khả năng tương thích của mô hình và trường hợp sử dụng

Not applicable; this is a tool for browsing and comparing models, not executing them.

Ghi chú về giấy phép và rủi ro

MIT license.

Editorial verification 2026-08-02: repository URL, owner, description, license and repository statistics were reviewed. License metadata: MIT. README was fetched for the channel draft; re-check repository dependencies, releases and model terms before production use.

Phát hành và bảo trì

v0.14.0 introduced richer model details (description, structured output, reasoning controls), deeper pricing (thinking-token, audio, tiered prices), a field glossary in the Models tab, and CLI parity for new fields. v0.13.0 added in-app custom agent addition and self-updates for agents. v0.12.1 added mouse support. v0.12.0 introduced multi-source benchmarks, a column picker, comparator columns, and prebuilt Nix binaries.

Firecrawl

firecrawl/firecrawl

★ 161,1KTypeScript

LangChain

langchain-ai/langchain

★ 143,6KPython

RAGFlow

infiniflow/ragflow

★ 86,7KGo