PBenchBrowse 284 families
PBench
Can the model plan, use tools, and finish a multi-step task?
Version / measure
- Domain
- Agents
- Ranked models
- 11
- Reported values
- 11
- 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 | Veo 3Google | reportedcost n/a1 value · Jun 15 · source | 0.526 |
| 2 | KlingExternal baseline | reportedcost n/a1 value · Jun 15 · source | 0.521 |
| 3 | WowExternal baseline | reportedcost n/a1 value · Jun 15 · source | 0.517 |
| 4 | LVPExternal baseline | reportedcost n/a1 value · Jun 15 · source | 0.515 |
| 5 | Wan2.6External baseline | reportedcost n/a1 value · Jun 15 · source | 0.514 |
| 6 | LTX-2External baseline | reportedcost n/a1 value · Jun 15 · source | 0.506 |
| 7 | VidarExternal baseline | reportedcost n/a1 value · Jun 15 · source | 0.501 |
| 8 | GigaWorldExternal baseline | reportedcost n/a1 value · Jun 15 · source | 0.495 |
| 9 | Sora2OpenAI | reportedcost n/a1 value · Jun 15 · source | 0.487 |
| 10 | CosmosNVIDIA | reportedcost n/a1 value · Jun 15 · source | 0.470 |
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. This benchmark reports points rather than percent correct.
Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.
PBench source-defined quality, consistency, motion, and domain metrics.
Method / source ↗