AGORA - Finance: leaderboard

Metric: Accuracy (%) on the 39 finance and economics questions (5,966 documents), archive-grounded agentic document reasoning with verifiable numeric answers over the domain collection of Markdown workplace documents, agent in the mini-swe-agent harness (bash tool only), temperature 1.0 with reasoning effort at the maximum each model supports; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 8 models tracked.

Top models

#ModelScore
1GLM-5.156.41
2GPT-5.5 (xHigh)53.85
3DeepSeek V4 Pro (Max)46.15
4Gemini 3.1 Pro (Preview) (High)41.03
5DeepSeek V4 Flash (Max)38.46
6Gemini 3.1 Flash Lite (High)5.13
7Qwen 3.5 35B A3B2.56
8Qwen 3.5 9B0

Interactive version: theaggregate.ai/benchmark?slug=agora-finance · How It Works · Data refreshed daily, snapshot 2026-09-29.