POST /v1/chat/completions
Web search lives at the top level of the request body in web_search_options — not inside the tools array. The two are mutually exclusive: a request with both returns 400.
Request
from openai import OpenAI
client = OpenAI(base_url="https://api.abliteration.ai/v1", api_key=os.environ["ABLIT_KEY"])
resp = client.chat.completions.create(
model="abliterated-model",
messages=[{"role": "user", "content": "Latest SEC filings for NVDA"}],
extra_body={
"web_search_options": {
"search_context_size": "medium",
"user_location": "us-east-1",
}
},
)
Options
| Field | Values | Effect |
|---|
search_context_size | "low" / "medium" / "high" | Number of results fetched (5 / 10 / 20). Default medium. |
user_location | string or location object | Hint for region-relevant results |
Domain restrictions
OpenAI’s Chat Completions spec does not define a native per-request domain filter for web_search_options. Domain restrictions come from one place:
Project-level allow/block via Policy Gateway. If your project has web_tools.allowed_domains or web_tools.blocked_domains set, the gateway enforces them server-side on every search made through this surface.
Configure them per project in the console under Project → Web tools. If you need per-request domain filters, use the Responses API or Anthropic Messages surfaces.Last modified on April 21, 2026
