GPT-5.3-Codex is OpenAI’s agentic coding model. It pairs the software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge of GPT-5.2. It scored at the top of SWE-Bench Pro and does well on Terminal-Bench 2.0 and OSWorld-Verified, which reflects better multi-language coding, terminal use, and real-world computer use. The model is tuned for long-running, tool-driven workflows and can be steered interactively during execution, which suits complex development work, debugging, deployment, and iterative product cycles. Outside coding, it also does well on structured knowledge-work benchmarks such as GDPval, with tasks like drafting documents, analyzing spreadsheets, creating slides, and operational research. It is trained with more cybersecurity awareness, including finding vulnerabilities, and has extra safeguards for higher-risk scenarios. Compared to earlier Codex models, it is more token-efficient and about 25% faster, and aims at end-to-end professional workflows that combine reasoning, execution, and computer interaction.
$0.875market price $1.75
$0.0875market price $0.175
$7.00market price $14.00
Throughput is how fast the model writes (tokens per second, higher is better). Time to first token is how long before the answer starts (lower is better).
Share of requests served without a server error. Client errors (4xx) do not count.
Works with the OpenAI SDK pointed at our base URL. Every parameter is in the API reference.
Daily usage of this model on NagaAI.
GPT-5.3-Codex is OpenAI’s agentic coding model. It pairs the software engineering performance of GPT-5.2-Codex with the broader reasoning and professional knowledge of GPT-5.2.
On NagaAI, GPT-5.3-Codex costs $0.875 per 1M input tokens and $7 per 1M output tokens. You pay only for what you use, with no subscription.
GPT-5.3-Codex reads up to 400k tokens in one request, counting the prompt and the answer together.
Yes. GPT-5.3-Codex can call your functions as tools and can answer in a JSON schema you give it.
It takes text and images as input and returns text.
Create an API key, copy the model ID from the top of this page and send requests with the OpenAI or Anthropic SDK pointed at NagaAI. The API section on this page has a ready example.
GPT-6.1 TerraOpenAIGPT-6.1 SolOpenAIGPT-6 SolOpenAIGPT-6 LunaOpenAIGPT-6 AstraOpenAIGPT 5.6 LunaOpenAINot enough throughput data