GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 series. It is suited for complex reasoning, coding, and agentic workflows, and is particularly strong at command-line and multi-step coding tasks...
Tested on 27 benchmarks · BenchGecko score 78.9. Top scores: OTIS Mock AIME 2024-2025 (100.0%), ARC-AGI (97.5%), GPQA diamond (95.1%).
Grok Build 0.1 scores 77.1 (98% as good) at $1.00/1M input · 50% cheaper
Unusual and adversarial machine learning challenges. Tests robustness of reasoning about edge cases in ML systems.
GitHub Star Optimization benchmark. Evaluates ability to understand and improve open-source projects to increase quality and adoption.
Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.
ARC-AGI 2, harder sequel to ARC. More complex abstract reasoning patterns that test generalization ability beyond training data.
Deceptively simple questions that humans find easy but AI models often get wrong. Tests common sense and reasoning gaps.
Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.
- Typemultimodal
- Context1.1M tokens (~525 books)
- ReleasedJul 2026
- LicenseProprietary
- StatusActive
- Cost / Message~$0.014
Frequently Asked Questions
Key facts · as of 2026-10-22
- GPT-5.6 Sol by OpenAI. BenchGecko score 78.9, rank 23 of 320 scored models (normalized average of public benchmark scores).
- List price $2.00 input · $10.00 output per 1M tokens (as of 2026-10-22).
- Sold by 7 providers (as of 2026-10-22): OpenAI $2.00 in / $10.00 out · Azure $4.00 in / $20.00 out · OpenAI $4.00 in / $20.00 out · Amazon Bedrock $4.40 in / $22.00 out · Azure $4.40 in / $22.00 out · and 2 more. Every provider
How to cite · data as of 2026-10-22
GPT-5.6 Sol · benchmarks, pricing and providers. BenchGecko, data as of 2026-10-22. https://benchgecko.ai/model/gpt-5-6-sol
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