工具调用让模型视觉推理得分近乎翻倍
Artificial Analysis 将生命科学视觉推理基准 VIALS-AA 改为无工具单次作答模式重跑,发现工具调用平均把模型得分从约 25% 提升到约 47%。Claude Fable 5.2 带工具以 67% 居首但单任务成本 4.20 美元,去掉工具后降至 44%;Claude Haiku 5.5 带工具得 49%、单任务仅 0.07 美元,反超前者无工具表现。结论是低成本模型加智能体框架可在成本与效果上同时胜过单次调用的大模型。
报道正文
Artificial Analysis (@ArtificialAnlys) · 2026-11-11 23
Take away its tools, and Claude Fable 5.2 scores below Claude Haiku 5.5 with tools on VIALS-AA; across models tool use nearly doubles scores, from ~25% to ~47% on average
VIALS-AA is our implementation of @joinHandshakeAI's VIALS benchmark for visual reasoning in life sciences research. By default, models work as agents: they think, run code and view images in a sandbox. We reran VIALS-AA in a one-shot format: one request with the question and images, and no tools.
Lower cost models in an agent harness can beat single-shot large models on both cost and performance in visual reasoning and interpretation of scientific artifacts.
Claude Fable 5.2 (max) leads VIALS-AA at 67% with tools, at a cost of $4.20 per task; without tools it scores 44% at $0.73 per task. Claude Haiku 5.5 (max) scores 49% with tools, at $0.07 per task: higher than Fable without tools, at about a tenth of the cost.
