GPT-5.6 Luna

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-5.6 Luna is an OpenAI GPT-5.6 model optimized for cost-sensitive workloads, using the Responses API for efficient text generation.

Model Info
Context Window ↗	1,050,000 tokens
Terms and License	link ↗
More information	link ↗
Request formats	Responses
Pricing	Short-context input (per 1M)$0.20 Short-context cached input (per 1M)$0.02 Short-context cache write (per 1M)$0.25 Short-context output (per 1M)$1.20 Long-context input (per 1M)$0.40 Long-context cached input (per 1M)$0.04 Long-context cache write (per 1M)$0.50 Long-context output (per 1M)$1.80

Usage

const response = await env.AI.run(
  'openai/gpt-5.6-luna',
  {
    input: 'Summarize the benefits of API rate limiting in three bullets.',
    max_output_tokens: 256,
  },
)
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-5.6-luna",
  "input": "Summarize the benefits of API rate limiting in three bullets.",
  "max_output_tokens": 256
}'

- **Protects system stability:** Prevents overload, reduces outages, and ensures predictable performance during traffic spikes.
- **Ensures fair access:** Stops individual users or applications from consuming disproportionate resources.
- **Improves security and cost control:** Helps mitigate abuse, brute-force attacks, and unexpected infrastructure usage.

{
  "id": "resp_0e481f0cb46bf357016a4fe99bd5fc8194b6ce2049aa65ab08",
  "object": "response",
  "created_at": 1783622043,
  "status": "completed",
  "background": false,
  "billing": {
    "payer": "developer"
  },
  "completed_at": 1783622044,
  "error": null,
  "frequency_penalty": 0,
  "incomplete_details": null,
  "instructions": null,
  "max_output_tokens": 256,
  "max_tool_calls": null,
  "model": "gpt-5.6-luna",
  "moderation": null,
  "output": [
    {
      "id": "msg_0e481f0cb46bf357016a4fe99c2e288194a6fd0ce3f5f88786",
      "type": "message",
      "status": "completed",
      "content": [
        {
          "type": "output_text",
          "annotations": [],
          "logprobs": [],
          "text": "- **Protects system stability:** Prevents overload, reduces outages, and ensures predictable performance during traffic spikes.\n- **Ensures fair access:** Stops individual users or applications from consuming disproportionate resources.\n- **Improves security and cost control:** Helps mitigate abuse, brute-force attacks, and unexpected infrastructure usage."
        }
      ],
      "phase": "final_answer",
      "role": "assistant"
    }
  ],
  "parallel_tool_calls": true,
  "presence_penalty": 0,
  "previous_response_id": null,
  "prompt_cache_key": null,
  "prompt_cache_retention": "in_memory",
  "reasoning": {
    "context": "all_turns",
    "effort": "medium",
    "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",
  "usage": {
    "input_tokens": 19,
    "input_tokens_details": {
      "cache_write_tokens": 0,
      "cached_tokens": 0
    },
    "output_tokens": 66,
    "output_tokens_details": {
      "reasoning_tokens": 0
    },
    "total_tokens": 85
  },
  "user": null,
  "metadata": {}
}

Examples

Pull Request Description — Using instructions for a cost-sensitive drafting task

const response = await env.AI.run(
  'openai/gpt-5.6-luna',
  {
    input:
      'Write a brief pull request description for a bug fix that prevents duplicate webhook deliveries.',
    instructions: 'Keep it under 100 words.',
    max_output_tokens: 256,
  },
)
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-5.6-luna",
  "input": "Write a brief pull request description for a bug fix that prevents duplicate webhook deliveries.",
  "instructions": "Keep it under 100 words.",
  "max_output_tokens": 256
}'

## Summary
Prevents duplicate webhook deliveries by ensuring each event is processed only once, even when retries or concurrent requests occur.

## Changes
- Added idempotency checks for webhook events.
- Prevented duplicate delivery attempts.
- Added regression tests covering retries and concurrent processing.

## Testing
All existing and new tests pass.

{
  "id": "resp_0bed05adeda548a1016a4fe99d1e408194bb0375bad96ab71a",
  "object": "response",
  "created_at": 1783622045,
  "status": "completed",
  "background": false,
  "billing": {
    "payer": "developer"
  },
  "completed_at": 1783622046,
  "error": null,
  "frequency_penalty": 0,
  "incomplete_details": null,
  "instructions": "Keep it under 100 words.",
  "max_output_tokens": 256,
  "max_tool_calls": null,
  "model": "gpt-5.6-luna",
  "moderation": null,
  "output": [
    {
      "id": "msg_0bed05adeda548a1016a4fe99daa748194992854423ba08e72",
      "type": "message",
      "status": "completed",
      "content": [
        {
          "type": "output_text",
          "annotations": [],
          "logprobs": [],
          "text": "## Summary\nPrevents duplicate webhook deliveries by ensuring each event is processed only once, even when retries or concurrent requests occur.\n\n## Changes\n- Added idempotency checks for webhook events.\n- Prevented duplicate delivery attempts.\n- Added regression tests covering retries and concurrent processing.\n\n## Testing\nAll existing and new tests pass."
        }
      ],
      "phase": "final_answer",
      "role": "assistant"
    }
  ],
  "parallel_tool_calls": true,
  "presence_penalty": 0,
  "previous_response_id": null,
  "prompt_cache_key": null,
  "prompt_cache_retention": "in_memory",
  "reasoning": {
    "context": "all_turns",
    "effort": "medium",
    "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",
  "usage": {
    "input_tokens": 33,
    "input_tokens_details": {
      "cache_write_tokens": 0,
      "cached_tokens": 0
    },
    "output_tokens": 69,
    "output_tokens_details": {
      "reasoning_tokens": 0
    },
    "total_tokens": 102
  },
  "user": null,
  "metadata": {}
}

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