Public Benefits Bench v1 - Both
Professional Work · 2026-06-09
The Both split of Public Benefits Bench v1 allows both web search and multi-turn interaction. It is kept under the archived v1 benchmark because v1.1 publishes a different leaderboard without these four condition splits.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5 | 71.7 |
| Opus 4.8 | 62.1 |
| MiniMax M3 | 60.7 |
| Sonnet 4.6 | 58.5 |
| Gemini 3.5 Flash | 58.0 |
| GLM-5.1 | 57.9 |
| DeepSeek-V4-Pro | 57.6 |
| GPT-5.5 | 57.2 |
| Kimi K2.6 | 53.4 |
| Gemini 3.1 Pro Preview | 50.9 |
| Grok 4.3 | 50.1 |
| Haiku 4.5 | 49.5 |
| Grok 4.1 Fast | 44.3 |