GPT-6 Astra

Model ID

Use with the Workers AI binding, the AI Gateway REST API, or the Unified API.

Calling the provider directly? Use the provider's model name instead, including with provider-native endpoints or the provider's own SDK. For example, Anthropic's name for anthropic/claude-opus-4.8 is claude-opus-4-8.

Third-party

GPT-6 Astra is OpenAI's most capable model, built for complex reasoning, coding, computer use, research, and document creation.

Model Info
Context Window ↗	1,050,000 tokens
Terms and License	link ↗
More information	link ↗
Request formats	Responses, Chat Completions
Pricing	Short-context input (per 1M)$10.00 Short-context cached input (per 1M)$1.00 Short-context cache write (per 1M)$12.00 Short-context output (per 1M)$50.00 Long-context input (per 1M)$20.00 Long-context cached input (per 1M)$2.00 Long-context cache write (per 1M)$25.00 Long-context output (per 1M)$75.00

Usage

const response = await env.AI.run(
  'openai/gpt-6-astra',
  {
    input:
      'A service has 99.9% monthly availability and just had 31 minutes of downtime. Has it exceeded the monthly error budget for a 30-day month? Show the calculation briefly.',
    max_output_tokens: 512,
    reasoning: { effort: 'high' },
  },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/v1/responses \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "openai/gpt-6-astra",
  "input": "A service has 99.9% monthly availability and just had 31 minutes of downtime. Has it exceeded the monthly error budget for a 30-day month? Show the calculation briefly.",
  "max_output_tokens": 512,
  "reasoning": {
    "effort": "high"
  }
}'

No. For a 30-day month:

- Total time: \(30 \times 24 \times 60 = 43{,}200\) minutes
- Error budget: \(43{,}200 \times (1 - 0.999) = 43.2\) minutes
- Remaining budget: \(43.2 - 31 = 12.2\) minutes

Assuming 31 minutes is the total downtime this month, it is **within the error budget**.

{
  "id": "resp_0f4dd4f38f19eb60016a99d9b6aaec87d1ab6bb0c344527fbc",
  "object": "response",
  "created_at": 1788467638,
  "model": "gpt-6-astra",
  "output": [
    {
      "id": "rs_0f4dd4f38f19eb60016a99d9b77c6087d1ab41b79ccdf5a833",
      "type": "reasoning",
      "content": [],
      "encrypted_content": "gAAAAABqmdm5bNo3TE240J65AqELJf45m_Forsch1ik0e0iiz5B-XZsvljWjY5PfmKXyel-6WvdoKmKulsRo3PX0fCqdNQFUaSYufMITZh61KSdsWBeeKRAwczgj8ihQBEbSF95MmTwkoQzsHX5SBitcf0n7CrMhXt8-LcE2tXqY309dnpgzAtjeI8H318LR1JyAWKsbver3s30gE1TRh0GC3YisTLrfACSclHB8_cOPA5QvEwKJNljm8R7eVDJx7LTFoB2Av93eaSD7l4fbAJzR_afL8K_Akj5NsjYaZVrsbPPxTYlYpqRhE0c-3C2U0oJ31-4UVa6_nT94QYtnDJ9YNjEdRC8sMkq6DwLhkbIIJGVNJGqHmWP3vQu3JCq9VDUtC3ss3FCkgWwhIV5tAFTFOtppnHenz1Xi2UYwaBNwaHWYJdrTBidNcpEQwPcIdSRAfOHv2r-ueFywhAZmj8Rx_GcJO_5_RWjvglhuNRBTbC92icwLx6wt4LWYBpLW-0itkHNtdy2cqsiKYSxWyMRzaEwyhW3wQo6jC9w9MRUG35wT3drTsuvnMDV5yhtzCtxq57eTRpIpCM9343ITl5DUwS-d7vouI7QH2MAIkZknxTO-_vVPYVQrQ4Th0g6Cb8CrGd-nHQ39xYKnPG7oUmvRobLSaGpKmQREzp-uoQeVK8dypZCl3Xd1uHSUO7w-HKsob-t5whmpE-ziHFaqciWh9YcssC0YE1ztbq0f9tGRy386ZzkkihD23sGR0khZukCdiGSw3Njakpe4AMWujU-t2BtomDJa6pdtLEuvG7p3MbIbHRoXkyK-9h74NgBY2jO7fk0BgItjpmqhcPHelCtMxIhP_OroanM0rCUjfQdqcSBVxwOd_hJ5wcNHmWdyf3gVvKEO83M4cj-sVljJ1afzFfhFsAQH5Zw-JTeBI1udknm3oy5lDwEzRoZpTcQs2Ix8lyUBtZ4MypqRLTnKLXzyij8G8XdgTIqwUkOHhMrm8Zc0_ndcrwPoVjwTq0NugRGJMDAxPKl-aqg6XvEfCYCUF2WYIu8jGo3KK6An8GJ5HFLyidgUslKGJtN2TJmKBOPwc0fXoiW4qyUQdVi2Hz23KBMOQE77jIF0cccZAWDToq67zPLVw9I4j-VKRpti-tOfcg9tz0wr4zJRi49uOqTw038Xn03VXzTolhb3GwwQkMZxWLjam3buEX6SLuS1A7XXZn3aKeLu6KhUin6_aR3jP0s2_Y50Mjv3x0EouaIKQJEINqYRgi9K2SKAxzRSKjGksdNGNcNRGs6cydkHSJF-f60p9HRBbBsN7duP8qm0K2URSJ1DX30=",
      "summary": []
    },
    {
      "id": "msg_0f4dd4f38f19eb60016a99d9b7d7f087d191f17156c41c2f32",
      "type": "message",
      "status": "completed",
      "content": [
        {
          "type": "output_text",
          "annotations": [],
          "logprobs": [],
          "text": "No. For a 30-day month:\n\n- Total time: \\(30 \\times 24 \\times 60 = 43{,}200\\) minutes\n- Error budget: \\(43{,}200 \\times (1 - 0.999) = 43.2\\) minutes\n- Remaining budget: \\(43.2 - 31 = 12.2\\) minutes\n\nAssuming 31 minutes is the total downtime this month, it is **within the error budget**."
        }
      ],
      "phase": "final_answer",
      "role": "assistant"
    }
  ],
  "status": "completed",
  "usage": {
    "input_tokens": 44,
    "output_tokens": 151,
    "total_tokens": 195,
    "input_tokens_details": {
      "cache_write_tokens": 0,
      "cached_tokens": 0
    },
    "output_tokens_details": {
      "reasoning_tokens": 41
    }
  },
  "access_programs": {
    "cyber": "standard"
  },
  "background": false,
  "billing": {
    "payer": "developer"
  },
  "completed_at": 1788467641,
  "error": null,
  "frequency_penalty": 0,
  "incomplete_details": null,
  "instructions": null,
  "max_output_tokens": 512,
  "max_tool_calls": null,
  "moderation": null,
  "parallel_tool_calls": true,
  "presence_penalty": 0,
  "previous_response_id": null,
  "prompt_cache_key": null,
  "prompt_cache_retention": "24h",
  "reasoning": {
    "context": "all_turns",
    "effort": "high",
    "mode": "standard",
    "summary": null
  },
  "safety_identifier": null,
  "service_tier": "default",
  "store": false,
  "temperature": 1,
  "text": {
    "format": {
      "type": "text"
    },
    "verbosity": "medium"
  },
  "tool_choice": "auto",
  "tool_usage": {
    "image_gen": {
      "input_tokens": 0,
      "input_tokens_details": {
        "image_tokens": 0,
        "text_tokens": 0
      },
      "output_tokens": 0,
      "output_tokens_details": {
        "image_tokens": 0,
        "text_tokens": 0
      },
      "total_tokens": 0
    },
    "web_search": {
      "num_requests": 0
    }
  },
  "tools": [],
  "top_logprobs": 0,
  "top_p": 0.98,
  "truncation": "disabled",
  "user": null,
  "metadata": {},
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

Examples

Migration Safeguards — Generate a concise answer through Chat Completions

const response = await env.AI.run(
  'openai/gpt-6-astra',
  {
    messages: [
      { role: 'user', content: 'List three practical safeguards for a production API migration.' },
    ],
    max_completion_tokens: 256,
  },
)
console.log(response)

curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/v1/chat/completions \
  --header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "openai/gpt-6-astra",
  "messages": [
    {
      "role": "user",
      "content": "List three practical safeguards for a production API migration."
    }
  ],
  "max_completion_tokens": 256
}'

1. **Validate compatibility before rollout.** Run contract and integration tests against real client behaviors. Keep breaking changes behind a new API version, and verify authentication, error formats, and data schemas.

2. **Roll out gradually with monitoring.** Start with internal traffic or a small canary group. Track error rates, latency, and critical business metrics; pause expansion if predefined thresholds are exceeded.

3. **Prepare and test rollback.** Keep the old API available and use a feature flag or traffic switch to revert quickly. Ensure database changes remain backward-compatible, and rehearse the rollback procedure.

{
  "id": "chatcmpl-EK8QDqH1XsfW6m8nEHXxUkrKGcWG5",
  "object": "chat.completion",
  "created": 1788467641,
  "model": "gpt-6-astra",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "1. **Validate compatibility before rollout.** Run contract and integration tests against real client behaviors. Keep breaking changes behind a new API version, and verify authentication, error formats, and data schemas.\n\n2. **Roll out gradually with monitoring.** Start with internal traffic or a small canary group. Track error rates, latency, and critical business metrics; pause expansion if predefined thresholds are exceeded.\n\n3. **Prepare and test rollback.** Keep the old API available and use a feature flag or traffic switch to revert quickly. Ensure database changes remain backward-compatible, and rehearse the rollback procedure.",
        "refusal": null,
        "annotations": []
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 16,
    "completion_tokens": 151,
    "total_tokens": 167,
    "prompt_tokens_details": {
      "cached_tokens": 0,
      "cache_write_tokens": 0,
      "audio_tokens": 0
    },
    "completion_tokens_details": {
      "reasoning_tokens": 23,
      "audio_tokens": 0,
      "accepted_prediction_tokens": 0,
      "rejected_prediction_tokens": 0
    }
  },
  "service_tier": "default",
  "system_fingerprint": null,
  "gatewayMetadata": {
    "keySource": "Unified"
  }
}

Parameters

▶input

one ofrequired

instructions

string

temperature

numberminimum: 0maximum: 2

max_output_tokens

numberexclusiveMinimum: 0

top_p

numberminimum: 0maximum: 1

stream

boolean

▶tools[]

array

tool_choice

▶text{}

object

▶reasoning{}

object

string

object

stringconst: response

created_at

number

model

string

▶output[]

array

output_text

string

status

stringenum: in_progress, completed, failed, incomplete

▶usage{}

object

API Schemas (Raw)

Input

Output