OpenAI
GPT-6.2 Astra
OpenAI's most capable GPT-6.2 model for the hardest software, computer-use, and document work
PositioningThe most capable GPT-6.2 model, for the hardest end-to-end projects
Context Length1,050,000 tokens
Max Output128K tokens
Pricing & Specs
๐ฐ Pricing
Input$8 / M tokens
Output$40 / M tokens
Cache Write$10 / M tokens
Cache Hit$0.40 / M tokens
โ๏ธ Specs
PositioningThe most capable GPT-6.2 model, for the hardest end-to-end projects
Context Length1,050,000 tokens
Max Output128K tokens
MultimodalText and image input; text output
API CompatibilityResponses API; Chat Completions API
Reasoning Effortlow, medium, high, xhigh, max; start new integrations at medium; none and minimal are unsupported
Tool CallingUse Responses API for tool calling; Chat Completions does not support tool calls
Short-context Pricing (โค272K input)Input $8; cached input $0.40; cache write $10; output $40 / M tokens
Long-context Pricing (>272K input)Input $16; cached input $0.80; cache write $20; output $60 / M tokens for the full request
Fast Mode2x the applicable token rates, including long-context pricing
Release DateNovember 17, 2026
Knowledge CutoffMay 18, 2026
API Examples
Python
from openai import OpenAI
client = OpenAI(api_key="YOUR_API_KEY", base_url="https://api.xairouter.com/v1")
response = client.responses.create(
model="gpt-6.2-astra",
input="Review this pull request and list the riskiest changes.",
)
print(response.output_text)
cURL
curl https://api.xairouter.com/v1/responses \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"gpt-6.2-astra","input":"Hello!"}'
Official references (verified November 18, 2026): GPT-6.2 Astra ยท OpenAI Pricing.