vava-nessa/free-coding-models

Find, benchmark and install in CLI 170+ FREE coding LLM models across 15+ providers in real time

What it solves

Finding the fastest and most reliable free AI coding model is difficult because there are hundreds of options across dozens of providers. It is hard to know which model is actually stable versus just temporarily fast, and manually configuring these models into your coding tools is tedious.

How it works

The tool pings 252 models from 24 providers in parallel to measure real-world performance. It calculates a "Stability Score" based on latency, jitter, and uptime, rather than just average speed. Once you select a model, the tool automatically updates your coding software's configuration so you can start working immediately.

Who it’s for

It is designed for developers who use AI coding assistants and want to optimize their workflow by using the best-performing free models without the hassle of manual configuration or constant performance testing.

Highlights

  • Live latency and stability rankings for 252 models across 24 providers.
  • One-click integration with popular tools like Aider, Continue, Cline, OpenHands, and more.
  • Includes a local "Smart Model Router" that provides an OpenAI-compatible endpoint with automatic failover.
  • Offers a terminal-based dashboard (TUI) and a web-based control center for monitoring and selecting models.
  • Features a "Smart Recommend" wizard to help you pick the right model based on your needs.
  • Includes AI speed tests and real-world telemetry scores to ensure you are getting the best performance.

Related

  • Project
    jgravelle/jcodemunch-mcp

    An MCP server that uses tree-sitter AST parsing to let AI agents retrieve precise code symbols instead of entire files, reducing token costs by up to 99%.

  • Project
    teaql/teaql-agent-kit

    TeaQL Agent Kit is a model-mediated harness for coding agents. It inserts an inspectable domain model between requirements and code: the agent models the business, evaluates and repairs the model, generates a typed API, and delivers a runnable app with audit evidence.

  • Project
    Cranot/claude-code-guide

    Claude Code is a terminal-based AI agent that understands your codebase to perform direct file edits, execute shell commands, and assist with development tasks through natural language.

  • Project
    runkids/skillshare

    skillshare is a Go‑based CLI (and optional desktop app) that stores, versions, audits and syncs AI‑coding “skills”, agents, rules and plugins across tools like Claude Code, Codex, Pi, OpenCode, etc. It offers a unified source of truth, security audits, multi‑tool sync, and a visual dashboard.