Versions

2 versions · Metric: Finance Agent Accuracy

Versions of Finance Agent Benchmark
VersionReleasedModelsResults
Finance Agent Benchmark 2 — October 2026 evaluationLatestOctober 6, 20265050
Finance Agent Benchmark 2 — September 2026 evaluationSeptember 30, 202611

Recently reported results: Finance Agent Benchmark 2 — October 2026 evaluation

Latest version: Finance Agent Benchmark 2 — October 2026 evaluation

Recently reported results for Finance Agent Benchmark 2 — October 2026 evaluation
ModelScoreSource
Kimi K2.644.9%vals.ai for Kimi K2.6 on Finance Agent Benchmark 2 — October 2026 evaluation (opens in a new tab)Archived copy (opens in a new tab)Finance Agent AccuracyReported Benchmark leaderboardVals AIIndependentRecord for Kimi K2.6 on Finance Agent Benchmark 2 — October 2026 evaluation
NVIDIA Nemotron 3 Ultra 550B-A55B37.7%vals.ai for NVIDIA Nemotron 3 Ultra 550B-A55B on Finance Agent Benchmark 2 — October 2026 evaluation (opens in a new tab)Archived copy (opens in a new tab)Finance Agent AccuracyReported Benchmark leaderboardVals AIIndependentRecord for NVIDIA Nemotron 3 Ultra 550B-A55B on Finance Agent Benchmark 2 — October 2026 evaluation
Qwen3.8-Max50.6%vals.ai for Qwen3.8-Max on Finance Agent Benchmark 2 — October 2026 evaluation (opens in a new tab)Archived copy (opens in a new tab)Finance Agent AccuracyReported Benchmark leaderboardVals AIIndependentRecord for Qwen3.8-Max on Finance Agent Benchmark 2 — October 2026 evaluation
Claude Haiku 4.5 (thinking)31.0%vals.ai for Claude Haiku 4.5 (thinking) on Finance Agent Benchmark 2 — October 2026 evaluation (opens in a new tab)Archived copy (opens in a new tab)Finance Agent AccuracyReported Benchmark leaderboardVals AIIndependentRecord for Claude Haiku 4.5 on Finance Agent Benchmark 2 — October 2026 evaluation
Inkling (0.99)46.6%vals.ai for Inkling (0.99) on Finance Agent Benchmark 2 — October 2026 evaluation (opens in a new tab)Archived copy (opens in a new tab)Finance Agent AccuracyReported Benchmark leaderboardVals AIIndependentRecord for Inkling on Finance Agent Benchmark 2 — October 2026 evaluation