FrontierChallenge (Claude Code): leaderboard
Apodex FrontierChallenge: agents complete specified scientific workflows across 97 tasks (materials, life science, chemistry and more) and deliver checkable artifact bundles; pass rate is the share of tasks with a native score of at least 99.9. Live leaderboard, models in the Claude Code scaffold; percent, higher is better.
Metric: Pass Rate (%). Source: apodexai.github.io. Saturation forecast: Around 2029. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok 4.6 | 20.6 | #29 |
| 2 | Kimi K3 | 17.5 | #35 |
| 3 | Claude Opus 5 | 17.5 | #16 |
| 4 | Qwen 3.8 Max | 15.5 | #34 |
| 5 | DeepSeek V4 Pro (0813) | 13.4 | #58 |
| 6 | DeepSeek V4 Flash (0731) | 12.4 | #84 |
| 7 | Gemini 3.7 Flash | 10.3 | #24 |
| 8 | Apodex 1.1 | 10.3 | #96 |
| 9 | Qwen 3.5 397B A17B | 4.1 | #154 |
| 10 | GLM-5.2 | 3.1 | #80 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=frontierchallenge-claude-code · How It Works · Data refreshed daily, snapshot 2026-11-05.