gpt-6-astra API — per-token pricing and code examples
Price per 1M tokens
Cache rates apply when the provider reports cache tokens.
68% offAbout this model
gpt-6-astra is served through OpenAI Chat Completions, OpenAI Responses, Anthropic Messages. You pay $3.20 per 1M input tokens and $16.00 per 1M output tokens. That is 68% below the official price of $10.00 and $50.00. Cache reads cost $0.32 per 1M tokens when the provider reports cached tokens. It is included in budget packs from $99.00 up.
When to reach for GPT-6 Astra
OpenAI calls GPT-6 Astra its most capable model for the most demanding work, and lists complex reasoning, coding, computer use, research and document creation as the uses it is meant for. It was released on September 3, 2026, described in OpenAI's changelog as built for the hardest end-to-end work. Its knowledge cutoff is April 30, 2026.
According to the Astra model page, the context window holds 1,050,000 tokens, up to 922,000 of them as input, and a single response can run to 128,000 output tokens. GPT-6 Astra reads text and images and writes text. Fine-tuning is not offered. Astra fits best where one strong answer matters more than speed: an architecture review over a large codebase, a long research brief, or an agent that has to carry a long task from start to finish.
Effort, sampling and tools on Astra
OpenAI's Astra page says reasoning.effort supports low, medium, high, xhigh and max. The changelog adds that GPT-6 Astra does not support the none effort level, so there is no way to run it without reasoning. Set the effort explicitly on each Astra request so you know which level you are paying for.
Sampling controls are also closed. OpenAI states that GPT-6 Astra does not support custom temperature or top_p values or log probabilities. Remove those fields when you point an existing client at Astra.
OpenAI documents tool calling for GPT-6 Astra on its Responses API. On OpenAI's Chat Completions API you can still get text answers, but function calling is not available with this model. On Cent AI, the /v1/responses format is available for Astra.
{
"model": "gpt-6-astra",
"reasoning": { "effort": "xhigh" },
"input": "Review this service for data races and write a migration plan"
}
That is the /v1/responses shape. On /v1/chat/completions, effort goes in reasoning_effort instead.
One more behaviour worth knowing: OpenAI's changelog says misalignment monitoring runs asynchronously for Astra and can pause a conversation for review. Plan for that pause in long agent runs.
What it costs on Cent AI
GPT-6 Astra is billed per token from your prepaid USD balance, at the four rates shown in this page's price box. Every Astra request reasons, and that reasoning is billed as output, so effort is the biggest lever you have on cost. A hard task at max usually spends far more reasoning tokens than the same task at low, even when the final answer has the same length.
The model page lists prompt caching as supported. Cent AI charges the cache read and cache write rates only for cached tokens that OpenAI reports in the response usage. Because OpenAI says changing effort mid-conversation on Responses keeps the prefix cache, you can start an Astra session at a lower effort and raise it for the hard step without losing the cached prefix.
Using it from your tools
GPT-6 Astra can be reached on /v1/chat/completions, /v1/responses and /v1/messages. Cent AI does not change your parameters silently. If a request includes reasoning_effort: "none", a custom temperature or function tools on Chat Completions, the model rejects it and it comes back as an upstream_error with the model's status code, for example 400; the fix is in your request body.
Specification
- Context window
- —
- Input and output
- Text
- Endpoints
- /v1/chat/completions, /v1/responses, /v1/messages
- Smallest pack
- $99.00
Latency and error rate per model are on the status page. status.cent-ai.dev
What a request costs
Open the full calculatorCost per request
Estimate
gpt-6-astraAn estimate at price version 5. Your bill comes from the usage the provider reports for each request.
Save $1.619Code
curl https://api.cent-ai.dev/v1/chat/completions \
-H "Authorization: Bearer $CENT_AI_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-6-astra", "messages": [{"role": "user", "content": "Hello"}]}'Set CENT_AI_KEY to a key from the dashboard. OpenAI SDKs use https://api.cent-ai.dev/v1; Anthropic SDKs use https://api.cent-ai.dev.
Price history
Version 5, in effect since Nov 19, 2026.
Effective
per 1M tokens
Inputold $1.00 new $3.20Cache readold $0.394 new $0.32Cache writeold $1.00 new $4.00Outputold $5.00 new $16.00Effective
per 1M tokens
Inputold $1.00 new $1.00Cache readold $0.10 new $0.394Cache writeold $1.00 new $1.00Outputold $5.00 new $5.00Effective
per 1M tokens
Inputold $0.50 new $1.00Cache readold $0.05 new $0.10Cache writeold $0.50 new $1.00Outputold $2.50 new $5.00Effective
per 1M tokens
Inputold $0.50 new $0.50Cache readold $0.05 new $0.05Cache writeold $0.625 new $0.50Outputold $2.50 new $2.50
Compare with
- deepseek-v4.1-flash-uncensoredInput$0.156
- claude-sonnet-5.5Input$0.58
- claude-opus-5Input$1.15
- claude-fable-5Input$2.80
Questions
How much does gpt-6-astra cost on Cent AI?
$3.20 per 1M input tokens and $16.00 per 1M output tokens. Cache reads cost $0.32 and cache writes $4.00 per 1M tokens. Each request is rounded up once.
Is there a minimum charge?
A successful request costs at least $0.00002. A request that ends early, because the client left or the output was cut off, pays only for its tokens.
How do I pay?
Top up a prepaid USD balance with QRIS or USDT/USDC, or buy a budget pack once. Nothing renews automatically.
What if the provider does not report usage?
The request is billed from an estimate and marked estimated on the Requests page. An estimate never contains cache tokens.
Which endpoint do I call?
/v1/chat/completions, /v1/responses, /v1/messages. The OpenAI base URL is https://api.cent-ai.dev/v1.
Try gpt-6-astra with one key
One prepaid balance for every model. Pay with QRIS or USDT/USDC.