AlgoWorlds - Feasibility: leaderboard
AlgoWorlds (2026) turns combinatorial optimization problems from ten families (such as routing, siting, dispatch and portfolio selection) into 240 partially observed environments where an agent gathers information through tools and then commits one decision. Feasibility is the share of worlds where the committed decision satisfies every constraint, in percent over three trials; higher is better.
Metric: Feasibility (%; share of worlds where the committed decision satisfies every task constraint; 240 partially observed environments built from ten combinatorial optimization families at four workload levels; the agent sees a hidden instance only through task-specific information tools, then commits one structured decision checked by an independent verifier; every model at its maximum reasoning setting; mean of three trials). Source: arxiv.org. Saturation forecast: Estimated already saturated. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.6 Sol (Max) | 97.5 | #29 (GPT-5.6 Sol) |
| 2 | Claude Opus 4.8 (Max) | 96.25 | #56 (Claude Opus 4.8) |
| 3 | Claude Sonnet 5 (Max) | 92.08 | #108 (Claude Sonnet 5) |
| 4 | GPT-5.6 Terra (Max) | 87.78 | #60 (GPT-5.6 Terra) |
| 5 | GLM-5.2 (Max) | 64.58 | #137 (GLM-5.2) |
| 6 | Qwen 3.5 Plus | 48.89 | #227 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=algoworlds-feasibility · How It Works · Data refreshed daily, snapshot 2026-11-12.