GPT-5.5
By OpenAI · released 2026-04-23 (6 months) · 5 effort levels
3 of 5 on the frontier at launchOff the frontier todayproprietary
- Intelligence index
- 33.80–38.40
- Cost to run the AA index
- $2095–$5294
- Price per 1M tokens (in / out)
- $5.00 / $30.0
Today
Where GPT-5.5 stands on the 2026-10-24 frontier
No measured level is on the frontier today.
| Level | Input modalities | Intelligence | Cost | E2E | Today | Matchup |
|---|---|---|---|---|---|---|
| GPT-5.5 (Medium) | Text, Image | 33.80 | $2095 | 9.73s | beaten by GPT-6.1 Sol (Max) | Compare |
| GPT-5.5 (High) | Text, Image | 37.00 | $3411 | 19.8s | beaten by Kimi K3.1 Pro | Compare |
| GPT-5.5 (Xhigh) | Text, Image | 38.40 | $5294 | 45.0s | beaten by Claude Opus 5.5 (Xhigh, Default Fallback) | Compare |
| GPT-5.5 (Low) | Text, Image | 30.70 | - | 8.14s | no data | Compare |
| GPT-5.5 (Non-reasoning) | Text, Image | 23.20 | - | 7.67s | no data | Compare |
3 of 5 effort levels plotted. The remaining levels lack intelligence or benchmark cost data.
GPT-5.5 effort levelsToday's Pareto frontierEvery other model
Matchup
Compare head to head
Suggested: what beats it, its effort levels, and frontier models near its price.
- GPT-5.5 (Xhigh) vs Claude Opus 5.5 (Xhigh, Default Fallback)
- GPT-5.5 (Xhigh) vs Claude Opus 5.5 (Max, Default Fallback)
- GPT-5.5 (Xhigh) vs GPT-5.5 (Medium)
- GPT-5.5 (Xhigh) vs GPT-5.5 (High)
- GPT-5.5 (Xhigh) vs GPT-5.5 (Low)
- GPT-5.5 (Xhigh) vs GPT-5.5 (Non-reasoning)
- GPT-5.5 (Xhigh) vs Claude Opus 5.5 (High, Default Fallback)
- GPT-5.5 (Xhigh) vs Kimi K3.1 Pro
Launch day
What GPT-5.5 did to the frontier on 2026-04-23
Reached the frontier
GPT-5.5 (Medium) · GPT-5.5 (High) · GPT-5.5 (Xhigh)
Reached the frontier3 of 5
Impact (hypervolume gained)7.04
Rank, past year's releases#3 of 46
Smartest level38.40
Cheapest level$2095
3 of 5 effort levels plotted. The remaining levels lack intelligence or benchmark cost data.
Reached the frontierMissed itDisplaced from the old frontierFrontier the day beforeFrontier after the releaseRegion gained (impact)
Agents:
Accept: text/markdown or ?format=md returns this page as Markdown. Markdown: /releases/gpt-5-5?format=md · llms.txt · agent skill