Router One published a guide on 18 November 2026 saying it now lists OpenAI's GPT-6.2 Sol, which OpenAI released on 17 November.
- Model ID:
openai/gpt-6.2-sol. The gateway also accepts OpenAI's own id,gpt-6.2-sol. - Context: 1,050,000 tokens (up to 922,000 input and 128,000 output tokens, per OpenAI's model page).
- Pricing: two rate lines. Requests with more than 272,000 input tokens use the long-context line for the whole request, output included. Router One's catalog shows $0.40 input / $2.00 output per million tokens for the standard line.
- Endpoints:
/v1/responsesand/v1/chat/completions. According to the guide, OpenAI says tool calling requires the Responses API. The API rejects requests that set reasoning effort tonone. - Plans: per the guide, GPT-6.2 Sol sits in the Premium models tier of the Pro, Max and Ultra plans. Calls draw on that shared allowance. Calls beyond it, or without a plan, bill per token to the wallet.
- Codex: the guide names GPT-6.2 Sol as Codex CLI's default model and explains the one-line switch.
The guide also lists what changes for requests written for GPT-6.1 Sol. It compares GPT-6.2 Sol with GPT-6.2 Astra, which OpenAI recommends for the most demanding tasks. This is Router One's own description of its catalog.