About GPT 6 Astra
GPT 6 Astra is OpenAI's GPT-6 generation reasoning model, built for software engineering, computer and browser use, science, and long multi-step professional workflows. OpenAI positions it as a step change over the GPT-5.6 family in staying on task across many steps, understanding intent, and completing agentic work end to end, and it reports strong results on coding, mathematics, and computer-use benchmarks.
It accepts text and images, supports configurable reasoning effort (low, medium, high, xhigh, and max), streaming, structured outputs, and function calling, and works with the web search, file search, code interpreter, computer use, and MCP tools. Fine-tuning is not available.
Astra is the first OpenAI model classified at the Critical level for cybersecurity under the Preparedness Framework, and High for biological and chemical capability. Requests in those domains can be refused by safety classifiers unless the organization is enrolled in a Trusted Access program, so it is not a fit for offensive security work without that enrollment. OpenAI also notes that its chain of thought is less monitorable than earlier models.
| Metric | Value |
|---|---|
| Context Length | 1,050,000 tokens |
| Max Output Tokens | 128,000 tokens |
| Knowledge Cutoff | Apr 30, 2026 |
| Multilingual | Yes |
Ready to build with GPT 6 Astra?
Try GPT 6 Astra in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.
curl -sSf -X POST https://hub.oxen.ai/api/ai/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $OXEN_API_KEY" \
-d '{
"model": "gpt-6-astra",
"messages": [
{
"role": "user",
"content": "Try sending a message."
}
]
}'API endpoint
See the API reference for request and response formats.
https://hub.oxen.ai/api/ai/chat/completions