StructuralOptimization
Task 44 / 47
ISCSO2023
ISCSO 2023 large 3D truss (~284 members) sizing under complex loads and constraints for minimum weight or cost. High-dimensional decisions stress scalable optimizers and constraint handling in competition-grade structural software.
Model leaderboard
Updated v1 results · 2026-09-15 · raw score, higher is better.
| # | Participant | Raw score | Medal |
|---|---|---|---|
| 1 | Claude Opus 4.6 | -16,477,799 | Gold |
| 2 | GLM-5 | -17,840,974 | Silver |
| 3 | Gemini 3.1 Pro Preview | -20,092,179 | Bronze |
| 4 | Grok 4.20 | -30,028,112 | — |
| 5 | Seed 2.0 Pro | -42,625,694 | — |
| 6 | DeepSeek V3.2 | -55,182,772 | — |
| 7 | Qwen3 Coder Next | -66,126,745 | — |
| 8 | GPT-5.4 | -77,813,243 | — |
Framework results · original paper
Historical results from the original evaluators · normalized score (0–100).
| # | Participant | Score |
|---|---|---|
| 1 | Claude Opus 4.6 + ABMCTS | 100.0 |
| 2 | Claude Opus 4.6 + OpenEvolve | 99.4 |
| 3 | GPT-OSS + ABMCTS | 97.8 |
| 4 | GPT-OSS + OpenEvolve | 48.5 |
| 5 | Claude Opus 4.6 + ShinkaiEvolve | 0.0 |
| 6 | GPT-OSS + ShinkaiEvolve | 0.0 |
Model results use the updated scores and medal thresholds. No valid score earns zero medal credit. Historical framework scores use the original evaluators and have not been updated alongside the model results.