Models
GPT-6 Astra
Model id: gpt-6-astra
- Agentic
- Frontier
- Coding
- Reasoning
- Vision
- Multimodal
- Long context
- Tools
- Research
OpenAI’s most capable model for hard reasoning, coding, research, and end-to-end work.
Best for
Ideal when the job is long, complex, and a mistake is expensive.
- Provider
- OpenAI
- Context window
- 1,050,000
- Release date
- 2026-09-03
- Reasoning effort
- lowmediumhighxhighmaxDefault: low
Pricing
Input
$15.4231 / 1M
Cached input
$1.5423 / 1M
Output
$77.1155 / 1M
Cache write
$19.2789 / 1M
Long-context rate (above 272,000 tokens)
Input
$30.8462 / 1M
Output
$115.6733 / 1M
Cached input
$3.0846 / 1M
Cache write
$38.5578 / 1M
Web search
Per successful search
$0.0154 / search
Billed on top of tokens when the model uses web search.
Amounts in Canadian dollars, converted at the Bank of Canada indicative rate (2026-09-21).
Performance
Milliseconds
Last 7 days
- Time to first byte (p95)
- 2,441 ms
- Output speed (p95)
- 36.3 tok/s
- Samples
- 3 over 7 days
Measured on real traffic at Pass egress — not a synthetic benchmark.
Activity
Tokens
—
—
—
—
—
Top 5 tools · last 7 days
Tool activity appears only when at least 2 workspaces have used this model recently.
Availability
Availability (%)
Last 7 days
- Operational probes
- 99.98% over 7 days
- Samples
- 9,210
Quick start
Call the API with a tonia_ key and the exact model id below.
curl
curl https://pass.tonia.ca:8443/v1/chat/completions \
-H "Authorization: Bearer $TONIA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-6-astra",
"messages": [
{
"role": "user",
"content": "Bonjour"
}
]
}'Python (OpenAI SDK)
import os
from openai import OpenAI
client = OpenAI(
base_url="https://pass.tonia.ca:8443/v1",
api_key=os.environ["TONIA_API_KEY"],
)
response = client.chat.completions.create(
model="gpt-6-astra",
messages=[{"role": "user", "content": "Bonjour"}],
)TypeScript (OpenAI SDK)
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://pass.tonia.ca:8443/v1",
apiKey: process.env.TONIA_API_KEY,
});
const response = await client.chat.completions.create({
model: "gpt-6-astra",
messages: [{ role: "user", content: "Bonjour" }],
});