Time Horizon Index (Vals AI) leaderboard
- Claude Opus 5.5 leads
- 91.3%
- models measured
- 13
- lead over #2
- +0.8 pts
- median
- 18.8%
Leader on Time Horizon Index (Vals AI) · read October 21
On October 21, 2026, 13 of our catalog's models had a Time Horizon Index (Vals AI) result, and Anthropic Claude Opus 5.5 led with 91.3%, ahead of OpenAI GPT-6 Astra at 90.5%.
Missions in Kerbal Space Program flown by an agent. Overall mission progress, as Vals AI publishes it.
Role in the index
Time Horizon Index (Vals AI) is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5Anthropic | 91.3% | 88.8 | $20.00 | |
| 2 | GPT-6 AstraOpenAI | 90.5% | 85.1 | $50.00 | |
| 3 | Claude Fable 5.1Anthropic | 63.3% | 84.2 | $50.00 | |
| 4 | Gemini 4 ArgonGoogle | 52.5% | 81.0 | $10.00 | |
| 5 | GPT-5.6 SolOpenAI | 23.8% | 73.5 | $20.00 | |
| 6 | Claude Opus 5Anthropic | 18.8% | 83.0 | $25.00 | |
| 7 | Gemini 3.8 FlashGoogle | 18.8% | 61.6 | $3.75 | |
| 8 | Claude Opus 4.8Anthropic | 11.8% | 65.0 | $25.00 | |
| 9 | Kimi K3Moonshot AI | 10.5% | 62.3 | $15.00 | |
| 10 | GPT-5.5OpenAI | 8.3% | 71.3 | $30.00 | |
| 11 | Grok 4.5xAI | 6.3% | 58.6 | $6.00 | |
| 12 | Grok 4.6xAI | 5.8% | 65.7 | $6.00 | |
| 13 | Muse Spark 1.1Meta | 1.7% | 70.3 | $4.25 |
Only models in our catalog are listed, and each appears once with its best published run. 136 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out.