Products
API Docs
A technical preview of the planned Simit API.
The endpoints are live; create your key in the API console.
Basics
| Base URL | https://simitgpt.com/api/v1 |
| Authentication | Authorization: Bearer <API_KEY> |
| Format | JSON compatible with OpenAI Chat Completions |
Chat completions
bash
curl https://simitgpt.com/api/v1/chat/completions \
-H "Authorization: Bearer $SIMIT_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "simit-flash",
"messages": [
{"role": "user", "content": "What is the longest river in Türkiye?"}
],
"stream": false
}'
Python (with the OpenAI SDK)
python
from openai import OpenAI
client = OpenAI(base_url="https://simitgpt.com/api/v1", api_key="SIMIT_API_KEY")
stream = client.chat.completions.create(
model="simit-flash",
messages=[{"role": "user", "content": "Tell me a short history of Istanbul."}],
stream=True,
)
for chunk in stream:
print(chunk.choices[0].delta.content or "", end="")
Models
| Model | Description |
|---|---|
simit-flash | Fast, economical general-purpose model; reasoning with "thinking": {"type": "enabled"} |
simit-pro | Our strongest reasoning model |
Prices and balance: API Platform. Streaming ("stream": true) uses the same SSE format as OpenAI.
Error codes
| HTTP | Meaning |
|---|---|
| 400 | Invalid request |
| 401 | Invalid or missing API key |
| 413 | Request too large |
| 429 | Rate limit exceeded. Honour the Retry-After header |
| 5xx | Temporary server error. Retry with exponential backoff |
Get a key: API Console