Agent benchmarks

GAIA agent benchmark

OpenGPT keeps 15 published rows attached to the exact GAIA benchmark version and environment shown below.

Tasks: 165
Published rows: 15
Source data date: Aug 6, 2026
Identity reviewed: Historical

Evidence scope

HAL verified snapshot · Top 15 published configurations · public validation set · 165 questions. public validation questions. General-assistant questions requiring reasoning, multimodality, browsing, and tools. Source-specific; no normalized step or token limit published.

How OpenGPT handles this evidence

OpenGPT preserves the publisher's metric, task scope, environment, version, configuration fields, and source date. It does not create a cross-benchmark overall Agent score.

Comparison boundaries

A higher result applies only to the displayed benchmark version and conditions.
Agent results reflect the full system, not the base model alone.
Missing cost, reliability, safety, or latency data remains unreported rather than being inferred.
HAL reproduced these public-validation rows and publishes scaffold, model, score, cost, run count, and traces. The source has paused adding new models, and it does not publish a normalized tool or step budget.

HAL verified snapshot · Top 15 published configurations · public validation set · 165 questions

Showing 15 of 15 published rows. Every result remains attached to the source benchmark and configuration.