GPT-6 Astra is OpenAI’s model for demanding reasoning, coding, and research. Access it through Flatkey with an OpenAI-compatible client.
Performance
Measured on real Flatkey traffic, with production monitoring.
Activity
This chart reflects live Flatkey requests for GPT-6 Astra and stays unreported until enough traffic is collected.
Pricing
The price cards above show current Flatkey catalog rates. OpenAI direct pricing uses a separate long-context tier: above 272K input tokens, input and cache rates are 2× and output is 1.5× for the full request. This is OpenAI’s direct rule; Flatkey settlement follows its live catalog and your account’s billing settings.
Use the same Flatkey balance and API key across image, video, audio, and text models.
OpenAI documents a 1,050,000-token context window, 128,000 maximum output tokens, image input, streaming, function calling, structured output, and Responses API tools.
Use Astra for difficult software, research, and analytical tasks where the highest-capability GPT-6 route is worth the higher token cost.
Plan around the documented 1,050,000-token context and the 272K input threshold before sending large codebases or document collections.
OpenAI documents web search, file search, code interpreter, computer use, MCP, and other tools through the Responses API; verify Flatkey route availability before relying on a tool.
Combine text input/output with supported image input while keeping unsupported audio and video inputs out of the request.
Compare
Astra is positioned as the highest-capability GPT-6 route. Compare the live catalog before choosing it over Sol or Luna for workloads where latency and cost matter more than maximum capability.
| Capability | GPT-6 Sol / GPT-6 Luna | GPT-6 Astra |
|---|---|---|
| Catalog model ID | Check the live catalog | gpt-6-astra |
| Flatkey endpoint | Check the selected model page | /v1/chat/completions |
| Official context | Check the selected model page | 1,050,000 tokens |
| Official maximum output | Check the selected model page | 128,000 tokens |
| Positioning | Lower-cost GPT-6 routes | Most capable GPT-6 route |
Why use Flatkey for GPT-6 Astra?
Keep model routing, token dimensions, the long-context threshold, and adjacent GPT-6 options visible before production traffic scales.
Set gpt-6-astra explicitly and keep /v1/chat/completions in the client configuration.
Review Flatkey catalog rates separately from OpenAI direct rates instead of mixing them into one headline price.
Estimate input size before the request crosses 272K tokens, where OpenAI's direct long-context tier changes the whole-request rate.
Compare Astra with adjacent GPT-6 routes under the same Flatkey key, usage, and limit controls.
API
Four ways in, all on the same key and the same model catalog. Pick one to see a runnable example.
curl https://router.flatkey.ai/v1/chat/completions \ -H "Content-Type: application/json" \ -H "Authorization: Bearer sk-***" \ -d '{"model":"gpt-6-astra","messages":[{"role":"user","content":"Say hello in one sentence."}]}'
FAQ
Answers about GPT-6 Astra pricing, the 272K long-context rule, model ID, endpoints, context window, modalities, and Flatkey access.
