OpenAI
VerifiedGPT-6 Astra
GPT-6 Astra handles demanding reasoning, coding, research, computer use, and document workflows. It accepts text and images and produces text, with a 1.05M-token upstream context window and up to 128K output tokens.
Reference pricing
Imported snapshot values per 1M tokens. Your customer portal is authoritative for actual kineticRouter billing.
| Component | Mirrored snapshot | Official reference |
|---|---|---|
| Input | $1.5/M | $10/M |
| Output | $7.5/M | $50/M |
| Cache read | $0.15/M | $1/M |
| Cache write | $1.875/M | $12.5/M |
API example
Use the OpenAI-compatible kineticRouter route with GPT-6 Astra.
curl https://api.kineticrouter.com/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $KINETICROUTER_API_KEY" \
-d '{
"model": "openai/gpt-6-astra",
"messages": [{"role": "user", "content": "Hello!"}]
}'About this model
GPT-6 Astra handles demanding reasoning, coding, research, computer use, and document workflows. It accepts text and images and produces text, with a 1.05M-token upstream context window and up to 128K output tokens.
Snapshot FAQ
Imported reference copy; verify runtime behavior before production use.
How is GPT-6 Astra pricing shown?＋
The table preserves standard reference rates captured on September 11, 2026. Long-context requests, service tiers, and tools can incur different charges. Your kineticRouter console remains authoritative for account rates and billed usage.
What are the context and output limits for GPT-6 Astra?＋
OpenAI documents a 1,050,000-token context window and up to 128,000 output tokens. The imported gateway catalog records 1,000,000 input tokens and 128,000 output tokens; these limits are shown separately.
How can I call GPT-6 Astra through kineticRouter?＋
Use model ID openai/gpt-6-astra with your kineticRouter API key and the OpenAI-compatible base URL https://api.kineticrouter.com/v1. Confirm that your key can access this model. Hosted tools require the Responses API; Anthropic-native kineticRouter routing remains planned.
Which capabilities does GPT-6 Astra support?＋
The reference includes text and image input, text output, reasoning, function calling, structured outputs, prompt caching, web search, and code interpreter. Check protocol and tool support for your kineticRouter account before using a feature in production.