Benchmarks
TheAgentCompany
Models scored
16
Best score
0.52
Metric
% Score
TheAgentCompany is an AI evaluation tracked by Epoch AI, with 16 scored model versions on record. The scores here are reported by third parties rather than produced by Epoch AI running the evaluation itself.
Scored models span June 2024 to September 2025. The best score recorded is 0.52, measured as % Score. Models come from Anthropic, Google DeepMind,Google, OpenAI, Meta AI, Alibaba, Google DeepMind, Amazon, DeepSeek.
Full record
- Score metric
- % Score
- Models scored
- 16
- Best score recorded
- 0.52
- Earliest model scored
- Jun 7, 2024
- Most recent model scored
- Sep 29, 2025
- Third-party reported
- Yes
Leaderboard
| 01 | DeepSeek-V3.2-Exp | DeepSeek | 0.52 |
| 02 | gemini-2.5-flash-preview-09-2025 | Google DeepMind | 0.52 |
| 03 | claude-sonnet-4-20250514 | Anthropic | 0.43 |
| 04 | claude-3-7-sonnet-20250219 | Anthropic | 0.4 |
| 05 | gemini-2.5-pro-preview-05-06 | Google DeepMind | 0.39 |
| 06 | claude-3-5-sonnet-20241022 | Anthropic | 0.34 |
| 07 | gemini-2.0-flash-001 | Google DeepMind,Google | 0.19 |
| 08 | gpt-4o-2024-11-20 | OpenAI | 0.17 |
| 09 | Llama-3.1-405B-Instruct | Meta AI | 0.14 |
| 10 | Llama-3.1-70B-Instruct | Meta AI | 0.13 |
| 11 | qwen2.5-72b-instruct | Alibaba | 0.12 |
| 12 | gemini-1.5-pro-002 | Google DeepMind | 0.08 |
| 13 | amazon.nova-pro-v1:0 | Amazon | 0.06 |
| 14 | qwen2-72b-instruct | Alibaba | 0.04 |
More benchmarks
SourceEpoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/benchmarks. Licensed under CC BY 4.0.
← All benchmarks