Great application of small language models by Meta.

They find that an 8B model picks tools better than a 402B model when trained for the job.

This work focuses on tool selection for agents with more than ten thousand tools.

At that scale, tool descriptions won't fit in context, and a retriever can only rank candidates without knowing which tools the instruction actually needs. Many systems hand this step to a proprietary LLM.

They train Qwen3 models on a Plan, Retrieve, Select workflow. The model splits the instruction into substeps, the retriever returns candidates for each one, and the model commits one tool per substep. Training is SFT on about 20K curated ToolBench instructions, then RL with a reward that scores how many gold tools are covered and at what rank.

On ToolBench G3, the 8B model goes from 36.2% after SFT to 41.5% after RL, above prompted Llama-4-Maverick.

The trained selector also works with different retrievers and on tool corpora from new domains.

https://arxiv.org/abs/2610.31483

Chat with Paper: https://academy.dair.ai/papers/training-small-language-models-for-transferable-tool-selection-2610.31483