OpenAI: GPT-5.3 Codex
openai/gpt-5.3-codex
Access GPT-5.3 Codex from OpenAI using Puter.js AI API.
Get Started// npm install @heyputer/puter.js
import { puter } from '@heyputer/puter.js';
puter.ai.chat("Explain quantum computing in simple terms", {
model: "openai/gpt-5.3-codex"
}).then(response => {
document.body.innerHTML = response.message.content;
});
<html>
<body>
<script src="https://js.puter.com/v2/"></script>
<script>
puter.ai.chat("Explain quantum computing in simple terms", {
model: "openai/gpt-5.3-codex"
}).then(response => {
document.body.innerHTML = response.message.content;
});
</script>
</body>
</html>
# pip install openai
from openai import OpenAI
client = OpenAI(
base_url="https://api.puter.com/puterai/openai/v1/",
api_key="YOUR_PUTER_AUTH_TOKEN",
)
response = client.chat.completions.create(
model="openai/gpt-5.3-codex",
messages=[
{"role": "user", "content": "Explain quantum computing in simple terms"}
],
)
print(response.choices[0].message.content)
curl https://api.puter.com/puterai/openai/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_PUTER_AUTH_TOKEN" \
-d '{
"model": "openai/gpt-5.3-codex",
"messages": [
{"role": "user", "content": "Explain quantum computing in simple terms"}
]
}'
Model Card
GPT-5.3 Codex is OpenAI's most capable agentic coding model, combining frontier coding performance with strong general reasoning and professional knowledge capabilities. It was the first model instrumental in creating itself, having been used to debug its own training and manage its own deployment. It sets state-of-the-art on SWE-Bench Pro and Terminal-Bench while being 25% faster than its predecessor.
Context Window 128K
tokens
Max Output 128K
tokens
Input Cost $1.75
per million tokens
Output Cost $14
per million tokens
Input text, image
modalities
Tool Use Yes
Knowledge Cutoff Aug 31, 2025
Release Date Feb 24, 2026
Output Speed 120
tokens / sec
Latency 31.50s
time to first token
Model Playground
Try GPT-5.3 Codex instantly in your browser.
This playground uses the Puter.js AI API — no API keys or setup required.
Benchmarks
How GPT-5.3 Codex performs on standard evaluations.
| Benchmark | Score |
|---|---|
| GPQA Diamond Graduate-level science Q&A | 91.5% |
| Humanity's Last Exam Cross-domain reasoning | 42.5% |
| IFBench Instruction following | 75.4% |
| LCR Long-context reasoning | 83.3% |
| Terminal-Bench Hard Agentic terminal tasks | 53.0% |
| τ²-Bench Tool use / agents | 86.0% |
Scores sourced from Artificial Analysis.
Find other OpenAI models →
GPT-6 Astra
GPT-6 Astra is OpenAI's most capable model, released in September 2026 as the successor to GPT-5.6 Sol. It targets complex reasoning, coding, computer use, research, and document creation, taking text and image input over a 1,050,000-token context window and supporting tool calling. OpenAI reports Astra saturates FrontierMath Tier 4 at 97.6% and ARC-AGI-3 at 99.9%, and cuts its hallucination rate from 92% to 51% at max effort versus GPT-5.6 Sol. It is the first OpenAI model to reach the Critical cybersecurity level under OpenAI's Preparedness Framework, scoring 100% on ExploitBench versus 78.5% for GPT-5.6 Sol. Independent testing from Artificial Analysis puts Astra's Intelligence Index at 61.2, roughly level with GPT-5.6 Sol and behind Claude Fable 5.1's 65.7. Input costs $5 per million tokens and output $25 per million, 2.5 times GPT-5.6 Sol's rate.
ChatGPT-6 Astra Pro
GPT-6 Astra Pro is OpenAI's higher-performance configuration of GPT-6 Astra, offered in ChatGPT to Pro, Business, and Enterprise subscribers. OpenAI has not published what separates it from the base model, whether a higher reasoning-effort ceiling, more compute per response, or fewer usage limits. It shares the same 1,050,000-token context window, 128,000 max output tokens, and $10/$50 per million token input/output pricing as base GPT-6 Astra. That overlap suggests Pro may simply be a higher-effort configuration of the same underlying model, a pattern also seen with GPT-5.6 Sol Pro. Base GPT-6 Astra supports adjustable reasoning effort from low to max, function calling, and image input, with an April 30, 2026 knowledge cutoff. It scores 67 on the Coding Agent Index and 61 on the Intelligence Index, near GPT-5.6 Sol and a few points behind Claude Fable 5.1. Whether these figures apply to the Pro configuration specifically is undocumented.
ChatGPT-5.6 Sol
GPT-5.6 Sol is OpenAI's flagship model in the GPT-5.6 family, sitting above Terra and Luna as the most capable and most expensive of the three. It reached general availability on July 9, 2026. Sol is built for agentic coding, cybersecurity research, and long-horizon autonomous work. OpenAI reports it sets a new state of the art on Terminal-Bench 2.1, a benchmark for command-line workflows that require planning, iteration, and tool coordination, and describes it as the most capable model yet for cybersecurity tasks such as vulnerability research. The GPT-5.6 family adds a max reasoning effort setting, an ultra mode that coordinates subagents on complex tasks, and Programmatic Tool Calling in the Responses API. Sol carries a 1.05M token context window, 128,000 max output tokens, and a February 16, 2026 knowledge cutoff, making it best suited for teams that need the highest available capability and can absorb its premium per-token cost.
Frequently Asked Questions
You can access GPT-5.3 Codex by OpenAI through Puter.js AI API. Include the library in your web app or Node.js project and start making calls with just a few lines of JavaScript — no backend and no configuration required. You can also use it with Python or cURL via Puter's OpenAI-compatible API.
Yes, it is free if you're using it through Puter.js. With the User-Pays Model, you can add GPT-5.3 Codex to your app at no cost — your users pay for their own AI usage directly, making it completely free for you as a developer.
| Price per 1M tokens | |
|---|---|
| Input | $1.75 |
| Output | $14 |
GPT-5.3 Codex was created by OpenAI and released on Feb 24, 2026.
GPT-5.3 Codex supports a context window of 128K tokens. For reference, that is roughly equivalent to 256 pages of text.
GPT-5.3 Codex can generate up to 128K tokens in a single response.
GPT-5.3 Codex has a knowledge cutoff date of Aug 31, 2025. This means the model was trained on data available up to that date.
GPT-5.3 Codex accepts the following input types: text, image. It produces: text.
Yes, GPT-5.3 Codex supports tool use (function calling), allowing it to interact with external tools, APIs, and data sources as part of its response flow.
GPT-5.3 Codex scores 36.9 on the Artificial Analysis Intelligence Index, outperforming 88% of tracked models.
Yes — the GPT-5.3 Codex API works with any JavaScript framework, Node.js, or plain HTML through Puter.js. Just include the library and start building. See the documentation for more details.
Get started with Puter.js
Add GPT-5.3 Codex to your app without worrying about API keys or setup.
Read the Docs View Tutorials