Benchmarks · composite
Vals RSI Index v1.1
Can a model do the research that builds the next model? Autonomous LLM R&D scored against frozen published, human, and frontier-model references. Overall score in percentage points.
21 models · 21 results · updated 2026-10-15 · vals.ai ↗
Leaderboard
Opus 5.5 max
More from this benchmark
6 published measuresScore vs any published measure6 measures: sub-scores, splits, counts
Breakdown
darker is better within each column · click a column to sort| Model | |||||
|---|---|---|---|---|---|
| Gemini 4 Argon max | 158.0M | 1.6M | 629k | 127.2M | 0 |
| GPT-6 Sol max | 1622.0M | 1.4M | 592k | 1612.6M | 9.3M |
| GPT-6.1 Sol max | 202.6M | 1.9M | 1.5M | 192.2M | 9.8M |
| Grok 4.7 max | 191.6M | 926k | 542k | 170.6M | 0 |
All results
21 rows| # | Model | Effort | Vals RSI Index v1.1 |
|---|---|---|---|
| 1 | Opus 5.5 | max | 37.3% |
| 2 | Fable 5.1 | max | 36.1% |
| 3 | Opus 5 | max | 33.0% |
| 4 | Gemini 4 Argon | max | 30.6% |
| 5 | GPT-6 Sol | max | 28.1% |
| 6 | GPT-6 Astra | max | 27.1% |
| 7 | GPT-6.1 Sol | max | 26.6% |
| 8 | Grok 4.6 | max | 25.5% |
| 9 | Fable 5 | max | 24.3% |
| 10 | Opus 4.8 | max | 24.0% |
| 11 | GPT-5.6 Sol | max | 23.9% |
| 12 | GLM 5.3 | max | 22.9% |
| 13 | Qwen 3.8 Max | max | 21.8% |
| 14 | Kimi K3 | max | 20.9% |
| 15 | Grok 4.7 | max | 20.2% |
| 16 | Gemini 3.8 Flash | max | 19.9% |
| 17 | Muse Spark 1.3 | max | 19.6% |
| 18 | Opus 4.7 | max | 19.1% |
| 19 | Gemini 3.7 Flash | max | 18.3% |
| 20 | GPT-5.5 | xhigh | 16.5% |
| 21 | GPT-5.4 | xhigh | 15.5% |