AISE-BenchBrowse 284 families
AISE-Bench
Can the model plan, use tools, and finish a multi-step task?
Version / measure
- Domain
- Agents
- Ranked models
- 6
- Reported values
- 6
- Reports
- 1 · lab
- Direction
- Higher is better
- Status
- Version not specified
Rankings
Best reported score per model across effort settings
| # | Model | Best setting | Score |
|---|---|---|---|
| 1 | Gemini 3 ProGoogle | reportedcost n/a1 value · Aug 7 · source | 0.5495 |
| 2 | Qwen3-235B-A22BQwen | reportedcost n/a1 value · Aug 7 · source | 0.4778 |
| 3 | DeepSeek-V3.2DeepSeekOW | reportedcost n/a1 value · Aug 7 · source | 0.4729 |
| 4 | GPT-5.2OpenAI | reportedcost n/a1 value · Aug 7 · source | 0.4562 |
| 5 | GLM-4.7Z.aiOW | reportedcost n/a1 value · Aug 7 · source | 0.3510 |
| 6 | Claude-4.5Anthropic | reportedcost n/a1 value · Aug 7 · source | 0.3328 |
Effort curve
Score vs. reported cost. Lines connect complete sweeps only.
No cost-linked effort sweep for this version.
Definition and comparison boundarypublic methodology
Can the model plan, use tools, and finish a multi-step task?
Agentic execution across tools, environments, or long-running workflows. Higher is better. The value is a ratio on a 0–1 scale reported in this lab's table.
Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.
CAW model row; source prints GLM-4.7 as 0.3510.
Method / source ↗