GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon agentic tasks that involve computer and browser use.
Modalities
In / Out Price
$10 / $50per 1M
Context
1M
Released
Sep 4, 2026
Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick — Balanced (price + speed), Nitro (fastest), or Exacto (highest tool-calling accuracy).
The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.
Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).
Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.
Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.
Token volume and request traffic to this model over time.
Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.
| Provider | Input /M | Output /M | Cache read /M | Latency | Throughput | Uptime |
|---|---|---|---|---|---|---|
| $5.00 | $25.00 | $0.50 | 6.95s | 51 tps | |
 | $10.00 | $50.00 | $1.00 | 5.93s | 47 tps | |
| $10.00 | $50.00 | $1.00 | 4.19s | 39 tps | |
| $11.00 | $55.00 | $1.10 | 3.29s | 30 tps | |
| $20.00 | $100.00 | $2.00 | 2.03s | 26 tps |
Throughput
51tok/s
P50, best across providers
Latency
2.03s
P50, best provider
100.00%
99.59%
When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and customization options.
GPT-6 Astra is OpenAI's flagship model for demanding end-to-end work. It is suited for advanced analysis, software engineering, deep research, scientific work, and document creation, with particular strengths in long-horizon agentic tasks that involve computer and browser use.
GPT-6 Astra costs $10.00/M input tokens and $50.00/M output tokens, with separate rates for Cache Read at $1.00/M tokens, Cache Write at $12.50/M tokens and Web Search at $10.00/1K calls.
GPT-6 Astra has a 1,050,000 token context window. It supports up to 128,000 completion tokens.
Yes. GPT-6 Astra accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
GPT-6 Astra accepts files such as PDFs, images and text as input and returns text.
GPT-6 Astra is served by 2 providers on OpenRouter: OpenAI and Azure (US). Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.
GPT-6 Astra was released on September 4, 2026.