Skip to content
Pricing
OpenAI

gpt-5.3-codex-spark API

GPT-5.3-Codex-Spark was a research preview for real-time coding. The launch record says it was not available through the public API at launch; this is historical information.

MODALITIES
TextText
Available
INPUT / OUTPUT90% OFF
$0.175/ 1M
/
$1.4/ 1M
USD per million tokens
CACHED INPUT
$0.0175/ 1M
Platform price · USD
CONTEXT WINDOW
128,000tokens
Max output: 16,384 tokens

API Usage Examples

API documentation ↗
POSTGet API Key ↗
curl 'https://astralgames.ai/v1/chat/completions' \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $MODEL_API_KEY" \
  --data '{
  "model": "gpt-5.3-codex-spark",
  "messages": [
    {
      "role": "user",
      "content": "Explain how an HTTP request works in three concise steps."
    }
  ],
  "max_completion_tokens": 1024,
  "stream": false
}'

Set MODEL_API_KEY in your server environment. Requests use this model’s native API and your account pricing.

Docs: authentication and keys · Error handling

gpt-5.3-codex-spark pricing

Estimate one request, then choose a budget for your workflow.

UsageQuantityRateCost
tokens$0.175/ 1M$0.000175
tokens$0.0175/ 1M$0
tokens$1.4/ 1M$0.0007
tokens$1.4/ 1M$0
Cost per request$0.000875

Public USD estimate. Your account rate and final recorded usage determine the actual charge.

Model pricing details
SpecificationOur priceOfficial reference
Input$0.175 / 1M tokens$1.75 About 90% less than official
Cached input$0.0175 / 1M tokens$0.175 About 90% less than official
Output$1.4 / 1M tokens$14 About 90% less than official
Reasoning$1.4 / 1M tokens$14 About 90% less than official

gpt-5.3-codex-spark API · Examples

Original prompts adapted to documented capabilities; these are not vendor-generated outputs.

Official source ↗ · 2026-10-14

EXAMPLE PROMPT

Find boundary errors in this TypeScript function. Provide a minimal fix and tests for nulls, repeated calls and errors. Code: [paste code]

USER MESSAGE
01

Code repair

Validate a minimal code change against explicit boundary cases.

What Is gpt-5.3-codex-spark?

GPT-5.3-Codex-Spark was a research preview for real-time coding. The launch record says it was not available through the public API at launch; this is historical information.

Official model information ↗ · 2026-10-14

gpt-5.3-codex-spark Platform specifications

Specificationgpt-5.3-codex-spark
Model IDgpt-5.3-codex-spark
ProviderOpenAI
API/v1/chat/completions · /v1/responses
Context window128,000 tokens
Maximum output16,384 tokens
Declared capabilitiesSee the native API contract
Protocol & capability details

Protocol & capabilities

openai · gpt-5.3-codex-spark

Context limit
128000
Maximum output
16384

Compare gpt-5.3-codex-spark with related models

Specificationgpt-5.3-codex-sparkgpt-5.4GPT-5.5
ProviderOpenAIOpenAIOpenAI
Input → outputTextTextTextTextTextText
Selection basisCurrent modelSame openai protocol · Same billing unitsSame openai protocol · Same billing units
Context / max output128,000 / 16,3841,050,000 / 128,0001,050,000 / 128,000
Starting price$0.175 / 1M tokens$0.25 / 1M tokens$0.5 / 1M tokens

Choose by the input types, output settings and billing unit your workflow needs. Try the same prompt across models to compare results for your own use case.

Integrate gpt-5.3-codex-spark into your app

Keep the native request format and point your application to the platform API. Prepare a key with model access, then review status and charges in your usage records.

View API examples

gpt-5.3-codex-spark API · Common Questions

Everything you need to know about API access, pricing and model capabilities.

How is gpt-5.3-codex-spark billed?

Charges depend on the recorded token usage and your account rates. Input, output and supported cache or reasoning categories are listed separately in the pricing table. The calculator uses public USD prices; the final usage receipt determines the charge.

How do I get an API key?

Choose “Get API access” to open the integration guide. Sign in to your Astral Nexus account, prepare a key with access to this model and copy an API example.

Which API endpoints can I use?

Supported endpoints: /v1/chat/completions、/v1/responses. Choose an endpoint in the API examples to see its native request format.

Does the model support streaming?

Streaming is not declared for the current route. Use the response format supported by its API.

Can I use tools or structured output?

Tools: not declared. Structured output: not declared. Request fields follow the selected native endpoint.

Where can I check usage and charges?

Keep the request ID and open your account’s LLM usage records to inspect request status, token counts and charges. Prompt and answer content are not a conversation history in this view.

Explore LLM Use Cases

All LLM models ↗

Integration topics / Documentation

All documentation ↗