Enterprise interface interaction on WorkArena L2 (test)
9.7Success RateA3-Qwen3.5-9B
Evaluation results
| Method | Links | |
|---|---|---|
| A3-Qwen3.5-9BHarness=GenericAgent, Params=9B2026.04 | 9.7 | |
| Qwen3.5-9B (base)Harness=GenericAgent, Params=9B2026.04 | 2.2 |
| Method | Links | |
|---|---|---|
| A3-Qwen3.5-9BHarness=GenericAgent, Params=9B2026.04 | 9.7 | |
| Qwen3.5-9B (base)Harness=GenericAgent, Params=9B2026.04 | 2.2 |