OpenAI · GPT-5.6
GPT-5.6 Sol
OpenAI’s main reasoning model for complex knowledge work, research, coding and tasks that need more deliberate analysis.
Treat the effort control as the product. Run one real job at two levels and keep the cheapest level that survives review.
Use it for
- Conflicting evidence and multi-step analysis
- Complex coding or computer-use work
- Difficult research and design decisions
Skip the extra latency for simple edits, obvious lookups and work where the source, not deeper reasoning, is missing.
Effort without the jargon
A considered answer should be enough and you want a sensible default for non-trivial work.
The task has competing constraints, several steps or a costly error that justifies more thought.
A Pro plan user has a genuinely difficult task and will compare the gain against time and allowance.
Advanced recordCost, limits, privacy and operations
Use the ChatGPT reasoning controls for hard work, but treat API cost, limits and privacy as a separate implementation decision that this product article does not fully specify.
- API cost
- Depends on surfaceChatGPT access varies by plan; API pricing is not stated in the cited ChatGPT model record, so no token price is reproduced here.
- Context window
- Not disclosedNot disclosed in the cited record
- Maximum output
- Not disclosedNot disclosed in the cited record
- Knowledge cutoff
- Not disclosedNot disclosed in the cited record
- Recorded inputs
- Not enumerated in cited record
- Recorded output
- Not enumerated in cited record
Data handling is a product decision
Depends on surfaceOpenAI says API data is not used for training unless the organisation opts in; retention varies by endpoint and controls. ChatGPT has separate product and workspace settings.
Operational constraints
- Availability is still rolling out and depends on plan and workspace settings.
- Reasoning allowances and fallbacks vary; an automatic switch is not the same as manually selecting Sol.
- Country availability and configured inference residency are separate constraints.
Unknown means the cited official record does not disclose a safe value. It is not an estimate. Prices are provider list prices in USD where stated and can change before this record's review date.
Keep the evidence separate
OpenAI says Sol powers Medium, High and Extra High in ChatGPT and targets complex work across coding, research, science, computer use and design.
His early use found Sol clearly competent but not obviously better than Claude Fable for his complex coding work; he also warns that price per token misses differences in reasoning volume.
Read the independent source ↗No Human Bit result is claimed; use this as evaluation guidance only.
Not scheduled
No Human Bit result is claimed for this model yet.