Our small models, as tools for your AI agent

Add this one address to Claude Code, Codex, Cursor or any MCP client, and each Udgam Labs model below becomes a tool your agent can call. New models appear here as new tools.

Add the server

# Claude Code
claude mcp add --transport http udgamlabs https://mcp.udgamlabs.com/mcp

# Codex
codex mcp add udgamlabs --url https://mcp.udgamlabs.com/mcp

# Cursor: .cursor/mcp.json
{ "mcpServers": { "udgamlabs": { "url": "https://mcp.udgamlabs.com/mcp" } } }

Tools

• shruti_transcribeSpeech to text for Hindi, English and Hinglish, from a link to an audio file. Shruti, 29.5 MB.
• vemana_continueContinues Telugu text. Vemana, a 278M Telugu base model: it writes on, it does not answer questions.
• laghu_chatOne message to Laghu, a 152M English chat model, with thinking on or off.
• find_small_modelMeasured scores of 27 open language models up to 400M parameters, from the Small Model Board.
• list_udgam_modelsEvery Udgam Labs model, with its size, licence, how to run it and where it falls short.

The models run on free shared CPUs on Hugging Face, so a call can take a few seconds, and the first call after a quiet spell waits while a model starts. Your text, or the link to your audio, goes to the model's Space to produce the answer; we do not log inputs or answers, and we count calls without IP addresses (privacy). To keep your data on your own device, run the models yourself: udgamlabs.com/models.