> ## Documentation Index
> Fetch the complete documentation index at: https://docs.snapgen.org/llms.txt
> Use this file to discover all available pages before exploring further.

# GPT-6 Astra

> Call gpt-6-astra through SnapGen using Chat Completions or Responses.

`gpt-6-astra` is available through SnapGen's OpenAI-compatible text endpoints.
It is a GPT-6 tier for demanding reasoning, coding, and agent workloads. Use the exact model ID with your SnapGen API key.

## Pricing and protocols

| Property | Value |
| - | - |
| Model ID | `gpt-6-astra` |
| Pricing | [Live GPT-6 prices](https://snapgen.org/models/gpt-6) |
| Chat Completions | `POST /v1/chat/completions` |
| Responses | `POST /v1/responses` |
| Streaming | Supported with `stream: true` |
| Tools | Function tools in the selected protocol's format |
| Long context | Prompts over 272,000 tokens use the model's long-context rate |

Billing uses reported input, output, and cached-input tokens. Reasoning tokens
bill as output. Check the [live model page](https://snapgen.org/models/gpt-6) for
current rates, including long-context pricing.

<ParamField body="model" type="string" required>
  Set to `gpt-6-astra`.
</ParamField>

<ParamField body="messages" type="array" required>
  Ordered messages for Chat Completions. Use `input` for Responses instead.
</ParamField>

<ParamField body="max_completion_tokens" type="integer">
  Caps generated tokens, including reasoning, for Chat Completions. `max_tokens`
  is a compatibility alias; `max_completion_tokens` wins if both are sent.
  Responses uses `max_output_tokens` instead.
</ParamField>

<ParamField body="reasoning_effort" type="string">
  Sets reasoning effort for Chat Completions, for example `medium`. Responses
  uses `reasoning: { "effort": "medium" }` instead. Supported levels vary by model.
</ParamField>

## Chat Completions request

```bash theme={null}
curl https://api.snapgen.org/v1/chat/completions \
  -H "Authorization: Bearer $SNAPGEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-6-astra",
    "messages": [{"role": "user", "content": "Summarize this design decision in three bullets."}],
    "reasoning_effort": "medium",
    "max_completion_tokens": 2048
  }'
```

Read generated text from `choices[0].message.content`. Set `stream: true` to
receive SSE chunks; add `stream_options: { "include_usage": true }` to request
streamed usage. See [Chat Completions](/api-reference/chat-completions) for
function tools and response fields.

## Responses request

```bash theme={null}
curl https://api.snapgen.org/v1/responses \
  -H "Authorization: Bearer $SNAPGEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-6-astra",
    "input": "List three risks in this rollout plan.",
    "reasoning": {"effort": "medium"},
    "max_output_tokens": 2048
  }'
```

Read `output[].content[]` items whose type is `output_text`, or use the SDK
`output_text` helper. See [Responses](/api-reference/responses) for streaming and
tool calls.

<Note>
  SnapGen forwards each protocol to routes that serve it natively. Use
  [`GET /v1/models`](/api-reference/list-models) to confirm availability for your
  key. The unsuffixed IDs `gpt-6` and `gpt-6.1` are not substitutes for these tier IDs.
</Note>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.