Frontier builder · United States
OpenAI
Frontier IC25 · Edition 2026.3 · Evidence reviewed October 8, 2026 UTC
A hosted model and agent platform with published API specifications and pricing.
Selected for a consequential, inspectable general-purpose model contribution. The displayed model is a named representative, not a claim of latest or best performance.
US operating company legal address in San Francisco; not inference geography.
Selected model
GPT-6.1 Sol · gpt-6.1-sol
Standard first-party synchronous API; uncached text tokens, prompt ≤100K.
1,050,000 · Context window
Scope and limitations
- Reasoning tokens occupy context and are billed at output rate; listed max output includes generated reasoning.
- Over 272,000 input tokens the entire request is priced at 2x input and 1.5x output; these rates apply below that threshold.
- Fast processing is 2x Standard. Regional processing adds 10% where available. GPT-6.1 Sol EU data residency has no Fast mode.
- This model selection does not describe OpenAI's separate open-weight model catalog.
Original sources
Explore the full IC25 tool · Read the field guide in six languages