Meta and Carnegie Mellon study finds an RL-trained 8B model beats Llama-4-Maverick at tool selection

A Qwen3-8B model post-trained with reinforcement learning hit 41.5% on ToolBench G3, ahead of the roughly 402B-parameter Llama-4-Maverick at 30.6% A language model with 8 billion parameters just outscored one roughly fifty times its size. The task was picking the right software tool for a job. Researchers from Meta and Carnegie Mellon University report that a small Qwen3-8B model, post-trained with reinforcement learning, selected the correct tools 41.5% of the time on the ToolBench G3 benchmark. Meta’s own Llama-4-Maverick, a model of approximately 402 billion parameters, scored 30.6% under similar conditions. The work appears in an arXiv paper titled “Training Small Language Models for Transferable Tool Selection,” listed as arXiv:2610.31483. It was submitted on October 22, 2026. The problem it tackles is a practical one for AI agents.

BeCrypto News shows a short excerpt and links to the publisher. The full report belongs to Crypto Briefing.