AgentWorldBench Search
AgentWorldBench Search domain - open-ended rubric evaluation for language world models simulating search environments. Normalized to 0-100.
Category
General Agents
Max Score
100.0
Score Type
percent
Active
Yes
Model Scores
Scores for AgentWorldBench Search
| Model | Score | Date | Verified |
|---|---|---|---|
| Qwen-AgentWorld-35B-A3B |
92.88%
|
24.06.2026 | Verified |
| Qwen3.5-35B-A3B |
25.44%
|
24.06.2026 | Unverified |
| Qwen3.5-397B-A17B |
55.86%
|
24.06.2026 | Unverified |
| Qwen3.6-Plus |
21.94
|
24.06.2026 | Unverified |
| GPT-5.4 |
96.47%
|
24.06.2026 | Unverified |
| Claude Opus 4.8 |
83.12%
|
24.06.2026 | Unverified |
| Claude Opus 4.6 |
46.35%
|
24.06.2026 | Unverified |
| Gemini 3.1 Pro |
52.08%
|
24.06.2026 | Unverified |
| Claude Sonnet 4.6 |
43.14%
|
24.06.2026 | Unverified |
| DeepSeek-V4-Pro |
35.71%
|
24.06.2026 | Unverified |
| GLM-5.1 |
3.27%
|
24.06.2026 | Unverified |
| Kimi K2.6 |
34.89%
|
24.06.2026 | Unverified |
| MiniMax-M2.7 |
33.75%
|
24.06.2026 | Unverified |
| Qwen-AgentWorld-397B-A17B |
100.00%
|
24.06.2026 | Unverified |