Cheaper GPT-5.3 Codex Spark API Pricing in 2026
Updated 2026-06-23
GPT-5.3 Codex Spark is available through APIsRouter. Set your SDK Base URL to https://api.apisrouter.com/v1, use model ID gpt-5.3-codex-spark, and inspect usage, cost, and errors in the console.
| API base URL | https://api.apisrouter.com/v1 |
|---|---|
| Interface | OpenAI-compatible /v1 |
| Billing | Pay as you go |
| Use case | model API |
| Model ID | gpt-5.3-codex-spark |
| Display name | GPT-5.3 Codex Spark |
| Context | 1M |
| APIsRouter price | $1.4 input / $11.2 output per 1M tokens |
When to use this model
GPT-5.3 Codex Spark can be called through the same OpenAI-compatible APIsRouter key used for other text and image models. Keep the Base URL fixed and switch only the model field when testing alternatives.
- Useful for agent tools, coding workflows, internal applications, and API experiments.
- Request logs show token usage, cost, latency, and sanitized errors.
- Use the model catalog and pricing page as the source of truth for availability and billing.
OpenAI SDK example
Fill Base URL at /v1 and put the model ID in the model field.
from openai import OpenAI
client = OpenAI(
api_key="sk-...",
base_url="https://api.apisrouter.com/v1",
)
response = client.chat.completions.create(
model="gpt-5.3-codex-spark",
messages=[{"role": "user", "content": "Reply with one sentence if the connection works."}],
)
print(response.choices[0].message.content)FAQ
What is the model ID for GPT-5.3 Codex Spark?
Use gpt-5.3-codex-spark. Model IDs are case-sensitive, so copying from the model catalog is safest.
How is GPT-5.3 Codex Spark billed?
Billing follows the live pricing page and request logs. Token models are charged by input/output/cache usage; image models are charged per request.
Do I need a separate API key for GPT-5.3 Codex Spark?
No. One APIsRouter key can call available text and image models unless you explicitly restrict that key.