Helium Market Resolution v4: leaderboard
Deterministic option-chain arithmetic benchmark (v4.0) scoring models on a predeclared 120-item stratified sample of frozen market snapshots across time value, Black-Scholes delta, put-call parity, implied-volatility inversion, and P(ITM).
Metric: Mean Credit (%) over 120 frozen option-chain arithmetic items. Source: huggingface.co. Saturation forecast: Estimated already saturated. 16 models tracked. Updated 20 Nov 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Gemini 4 Hydrogen | 99.07 | #16 | |
| 2 | GPT-5.6 Sol | 99.03 | #36 | |
| 3 | Gemini 4 Carbon | 99.03 | #5 | |
| 4 | GPT-6.2 Astra | 99.03 | #4 | |
| 5 | GPT-6.2 Sol | 99.03 | #12 | |
| 6 | Grok 4.6 | 98.98 | #50 | |
| 7 | Gemini 3.1 Pro (Preview) | 98.95 | #101 | |
| 8 | Claude Opus 5.5 | 98.86 | #9 | |
| 9 | Qwen 4 Max | 98.83 | #29 | |
| 10 | Claude Fable 5.2 | 98.3 | #1 | |
| 11 | Grok 4.20 0309 | Reasoning | 97.91 | #238 |
| 12 | Claude Opus 4.7 | 96.03 | #85 | |
| 13 | Grok 4.3 | 91.52 | #338 | |
| 14 | Claude Sonnet 5.5 | 72.66 | #24 | |
| 15 | GPT-5.4 | 51.01 | #138 |
No result here: #3 Gemini 4 Neon, #10 GPT-6 Astra.
Interactive version: theaggregate.ai/benchmark?slug=helium-market-resolution-v4 · How It Works · Data refreshed daily, snapshot 2026-11-20.