All models
OpenAI
GPT-5.3-Codex
OpenAI's agentic coding specialist: long, tool-heavy development sessions, refactors and debugging, with 400K context. Strong at code, weak at creative writing and roleplay. GPT-6.1 Sol now handles coding at a similar price and does more besides.
MainstreamThe big labs' models with their own safety filters. Best for work, code and research.
Brilliant
Intelligence
~6.37 credits
Estimated price per message
400K tokens
Context window
Text, Images
Reads
GPT-5.3-Codex is a legacy model. Its successor, GPT-6.1 Sol, is faster or cheaper - and chats that used GPT-5.3-Codex move to it automatically.
See GPT-6.1 SolThe facts
Made by
OpenAI
Content tier
MainstreamThe big labs' models with their own safety filters. Best for work, code and research.
Intelligence
Brilliant
Estimated price per message
~6.37 credits≈ 15 messages for 1 €
Context window
400K tokens≈ 615 pages of text in one go
Reads
Text, Images
Writes
Text
Tool use
Yes - search, memory, code and more
Served via
OpenRouter, Unbottled AI
Status
Legacy
A typical message: 16K tokens in, 1.5K out. Short replies cost less. Models served via OpenRouter bill the real provider cost plus our markup, so actual charges vary. 1 credit = 0.01 €.
Compare GPT-5.3-Codex
Same facts, side by side.
Try GPT-5.3-Codex now.
No subscription. Free credits to start. The chat opens with GPT-5.3-Codex selected and a prompt ready - just send.