OpenAI's smartest model · GPT-5.5. Same speed as GPT-5.4. Plans, uses tools, checks its own work. Tops Terminal-Bench 2.0 (82.7%), GDPval (84.9%), ARC-AGI-2 (85.0%), CyberGym (81.8%). SOTA on AA Coding Index at half the cost.
Tested on 6 benchmarks · BenchGecko score 65.7. Top scores: ARC-AGI (95.0%), GPQA diamond (93.6%), browsecomp (84.4%).
Gemini 3.7 Flash scores 66.4 (101% as good) at $0.75/1M input · 85% cheaper
Complex terminal-based engineering tasks. Models must use command-line tools, navigate filesystems, and debug systems through shell interaction.
Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.
Graduate-level science questions written by PhD experts. Diamond subset contains questions where experts disagree, testing deep understanding.
- Typetext
- Context400K tokens (~200 books)
- ReleasedApr 2026
- LicenseProprietary
- StatusActive
- Cost / Message~$0.040
Frequently Asked Questions
Key facts · as of 2026-10-05
- GPT-5.5 by OpenAI. BenchGecko score 65.7, rank 55 of 312 scored models (normalized average of public benchmark scores).
- List price $5.00 input · $30.00 output per 1M tokens (as of 2026-10-05).
- Sold by 7 providers (as of 2026-10-05): OpenAI $2.50 in / $15.00 out · Azure $5.00 in / $30.00 out · OpenAI $5.00 in / $30.00 out · Amazon Bedrock $5.50 in / $33.00 out · Azure $5.50 in / $33.00 out · and 2 more. Every provider
How to cite · data as of 2026-10-05
GPT-5.5 · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-05. https://benchgecko.ai/model/gpt-5-5
Credit "Source: BenchGecko" with a link. Prices per provider and Gecko Tests are BenchGecko data (CC BY 4.0); benchmark scores keep their original source, listed in the JSON. JSON · llms.txt · MCP