GPT-6 Astra on QuickSilver Pro
GPT-6 Astra is OpenAI's GPT-6 flagship for demanding end-to-end work — long-horizon agentic coding, deep research, scientific analysis and document creation — with a 1M-token context, 128K max output and image input. QuickSilver Pro charges $10.00 input / $50.00 output per million tokens — OpenAI's published list price, with no QSP markup — on the same key and balance as the rest of the catalog.
At a glance
OpenAI's flagship for long-horizon agentic coding, deep research and document work, with a 1M-token context and image input.
Pricing comparison ($/1M tokens)
| Provider | Input | Output | vs QSP |
|---|---|---|---|
| QuickSilver Pro | $10.00 | $50.00 | — |
| OpenAI list price (OpenAI API) | $10.00 | $50.00 | same |
When to use
Use GPT-6 Astra when the task is the hard part: multi-hour agent runs over a large codebase, research that has to read and reconcile many sources, scientific and quantitative analysis, and long structured documents. It accepts text, images and files across a 1M-token context, supports tool calling, JSON Schema output and both Chat Completions and Responses, and reasoning — always on for this model — is tuned per request with `reasoning_effort`.
When to use something else
For routine chat, extraction and high-volume automation, GPT-5.6 Terra and GPT-5.6 Luna do the job at a fraction of the price, and GPT-5.6 Sol remains the value pick for frontier work at $1.60 / $8.00. A request whose prompt exceeds 200,000 tokens is billed at OpenAI's long-context rate — $20 input, $75 output and $2 cached input per million — so use retrieval or prompt compaction when the full window is not needed.
Quickstart (curl)
curl https://api.quicksilverpro.io/v1/chat/completions \
-H "Authorization: Bearer $QSP_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-6-astra",
"messages": [{"role": "user", "content": "Hello!"}]
}'OpenAI-compatible. One-line migration via base_url.
FAQ
For requests up to 200,000 prompt tokens it costs $10.00 per million input tokens, $50.00 per million output tokens, $1 per million cached-input tokens and $12.50 per million cache-write tokens — OpenAI's published list rates, with no markup. Above 200,000 prompt tokens the whole request bills at $20 input, $75 output and $2 cached input per million. Reasoning tokens bill as output tokens.
Reasoning is always on — it is mandatory on this endpoint and cannot be disabled (a request that tries to turn it off is rejected upstream). It is light on simple prompts: in our testing a one-line question spent about a dozen reasoning tokens. Use `reasoning_effort` (low, medium, high, xhigh) to trade depth against latency and cost. Reasoning tokens bill as output tokens and draw from your `max_tokens` budget before the visible answer, so set `max_tokens` generously on hard problems.
Yes. It accepts text, image and file input and supports streaming, function calling and JSON Schema structured output across a 1M-token context, on both `/v1/chat/completions` and `/v1/responses`.
Set base_url=https://api.quicksilverpro.io/v1, use your QSP API key, and set model="gpt-6-astra". Everything else in your existing OpenAI code stays the same, and usage.cost in the response tells you what the call charged.
GPT-6 Astra is OpenAI's newest flagship generation, positioned for the longest-horizon agentic and research work; GPT-5.6 Sol is the previous flagship and stays on QuickSilver Pro at $1.60 / $8.00, 20% below its list price. Both share the 1M-token context and the same OpenAI-compatible surface, so switching between them is a one-line model change.