Codex Mini is a retired AI model from OpenAI.

Codex Mini has a recorded input context length of 200,000 tokens.

The lowest base rates currently recorded across providers for Codex Mini are Input Text Tokens at $1.50 per 1M tokens; Output Text Tokens at $6.00 per 1M tokens; and Cache Read Tokens at $0.38 per 1M tokens. The pricing section shows every recorded provider, pricing plan, meter, and condition.

Codex Mini is available through the Phaseo API, with 1 active provider currently recorded. The providers section shows the routes and availability currently recorded by Phaseo.

Phaseo does not currently have enough active route metadata to confirm whether Codex Mini supports tool calling.

Phaseo does not currently have enough active route metadata to confirm whether Codex Mini supports structured outputs.

Phaseo currently tracks 2 benchmark results for Codex Mini. Review the benchmark section for scores, ranks, methodology context, and available sources.

Codex Mini accepts image and text inputs. It produces text output.

Codex Mini was released on 16 May 2025.