ASI-Bench (Claude Code) - B4 Goal + Distractors: leaderboard
ASI-Bench prompt band B4 (goal plus distractors) on 60 project-level tasks without external tool access (arXiv 2608.17271v1). Scientific Score macro-averaged over tasks and rounds, 0-100, higher is better. Claude Code harness runs only, from the Apexintelligence-AI/ASI-Bench-Leaderboard Space's app/page.tsx; efforts stay in the label.
Metric: Scientific Score (0-100). Source: huggingface.co. Saturation forecast: Around 2029. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 (Max) | 42.39 | #16 (Claude Opus 5) |
| 2 | GLM-5.3 (Max) | 35.01 | #44 (GLM-5.3) |
| 3 | Gemini 3.8 Flash | 34.52 | #30 |
| 4 | Qwen 3.8 Max (Max) | 34.46 | #34 (Qwen 3.8 Max) |
| 5 | Claude Opus 4.8 (Max) | 29.59 | #39 (Claude Opus 4.8) |
| 6 | Kimi K3 (Max) | 29.57 | #35 (Kimi K3) |
| 7 | GLM-5.2 (Max) | 27.27 | #80 (GLM-5.2) |
| 8 | GLM-5.3 Flash (Max) | 25.43 | #77 (GLM-5.3 Flash) |
| 9 | DeepSeek V4 Flash (Max) | 24.41 | #155 (DeepSeek V4 Flash) |
| 10 | DeepSeek V4 Pro (Max) | 18.94 | #112 (DeepSeek V4 Pro) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=asi-bench-claude-code-b4-goal-plus-distractors · How It Works · Data refreshed daily, snapshot 2026-11-05.