Tool use

Function-calling on the Anthropic-native endpoint. The API normalizes tool_choice so legacy SDKs that emit {type:any} or {type:tool} keep working.

Request shape

Declare tools as an array. Each tool has a name, a description, and a JSON-Schema input_schema. The model emits a tool_use content block whenever it decides to call one.

{
"model": "claude-sonnet-5-5",
"max_tokens": 1024,
"tools": [
{
"name": "get_weather",
"description": "Return current weather for a city.",
"input_schema": {
"type": "object",
"properties": {
"city": { "type": "string" },
"units": { "type": "string", "enum": ["c", "f"] }
},
"required": ["city"]
}
}
],
"messages": [
{ "role": "user", "content": "What's the weather in Berlin in celsius?" }
]
}

Response with a tool call

{
"id": "msg_01...",
"stop_reason": "tool_use",
"content": [
{ "type": "text", "text": "Let me check that for you." },
{
"type": "tool_use",
"id": "toolu_bdrk_01...",
"name": "get_weather",
"input": { "city": "Berlin", "units": "c" }
}
]
}

Continuing the loop

After your code executes the tool, send the result back as a user message containing a tool_result block. The model then either emits another tool call or a final text answer.

{
"role": "user",
"content": [
{
"type": "tool_result",
"tool_use_id": "toolu_bdrk_01...",
"content": "{\"city\":\"Berlin\",\"temp_c\":18,\"sky\":\"clear\"}"
}
]
}

tool_choice — how we normalize

Anthropic accepts four tool_choice forms. The OneProvider API accepts all four from your client, but passes two of them to the model in a different form:

Parallel tool calls

A single response can contain multiple tool_use blocks — the model decides whether to fan out. Execute all of them, then send back one user message with multipletool_result blocks (one per tool_use_id). Order doesn’t matter.

Tool naming

• Names are case-sensitive and pass through verbatim — the model echoes whatever you registered.
• Both snake_case and PascalCase work. The Claude Code CLI uses PascalCase (Read, Bash, Edit); generic SDKs typically use snake_case.
• Keep names <64 chars and ASCII [a-zA-Z0-9_].

OpenAI-style tool calling

On the /v1/chat/completions endpoint, use the standard OpenAI tools /tool_calls shape — we translate it to Anthropic tool_use blocks under the hood.

Quick answers

Which base URL do I use?

Claude Code and Anthropic SDKs use https://api.oneprovider.dev. OpenAI SDKs and OpenAI-compatible clients use https://api.oneprovider.dev/v1. One key works behind both; the model list is on the models page ; rates are on the pricing page .

How do I authenticate?

Send the raw sk-… key in x-api-key or Authorization: Bearer: one of them, never both. Keep the key in an environment variable, not in a file that gets committed. More on authentication .

Where do I get a key?

Open your cabinet in the Telegram bot @oneprovider_robot or at app.oneprovider.dev, top up the balance and press “Issue a key”. Each key can carry its own spend limit. Cabinet & API keys .

What do 401, 404 and 429 mean?

401: the key is missing, mistyped or sent in the wrong header; paste the whole key without quotes or spaces. 404: usually a wrong base URL, for example a missing or doubled /v1. 429 with “Insufficient balance”: the cabinet balance or the key’s spend limit is used up, so top up instead of retrying; any other 429 is a rate limit, so wait for Retry-After. Details: error codes and troubleshooting .

Why do streams cut off or reset?

VPN and VLESS proxies often buffer SSE or break HTTP/2 negotiation. OneProvider has no regional restrictions and does not buffer streams, so turn the proxy off and retry.

Where do I ask if this is not enough?

Message @Oneproviderhelpbot on Telegram or open “Support” in the web cabinet, with the HTTP status code and the exact error message. How support works .