Chat Completions

POST https://api.smallest.ai/waves/v1/chat/completions

OpenAI-compatible chat completion endpoint. Most OpenAI Chat Completions request fields are accepted and passed through to Electron verbatim. The response shape matches OpenAI’s chat.completion object.

Request

Authentication

1
Authorization: Bearer $SMALLEST_API_KEY

Required fields

Common optional fields

Unrecognized parameters not in the supported set are accepted by the schema but may be ignored by the model. See Supported Parameters for the full passthrough table.

Explicitly rejected fields

Sending either returns HTTP 400 with invalid_request_error.

Response

Non-streaming (default)

Standard OpenAI chat.completion object:

1
{
2
  "id": "chatcmpl-a670053f3ab2ed8f",
3
  "object": "chat.completion",
4
  "created": 1740000000,
5
  "model": "electron",
6
  "choices": [
7
    {
8
      "index": 0,
9
      "message": {
10
        "role": "assistant",
11
        "content": "Hi! How can I help you today?",
12
        "tool_calls": null
13
      },
14
      "finish_reason": "stop"
15
    }
16
  ],
17
  "usage": {
18
    "prompt_tokens": 20,
19
    "completion_tokens": 9,
20
    "total_tokens": 29,
21
    "prompt_tokens_details": {
22
      "cached_tokens": 0
23
    }
24
  }
25
}

usage block

finish_reason

Streaming

When stream: true, the response is text/event-stream SSE. See Streaming for chunk format and example consumer code.

Errors

All errors follow the OpenAI-style envelope. Validation failures additionally include a details array with per-field reasons:

1
{
2
  "error": {
3
    "message": "Invalid input data",
4
    "type": "invalid_request_error",
5
    "details": [
6
      { "code": "custom", "message": "n > 1 is not supported", "path": ["n"] }
7
    ],
8
    "request_id": "9db74058-2421-4793-8796-b2629ff3eb1c"
9
  }
10
}

Examples

Minimal

1
import os
2
from openai import OpenAI
3
4
client = OpenAI(
5
    base_url="https://api.smallest.ai/waves/v1",
6
    api_key=os.environ["SMALLEST_API_KEY"],
7
)
8
9
resp = client.chat.completions.create(
10
    model="electron",
11
    messages=[
12
        {"role": "system", "content": "You are a concise assistant."},
13
        {"role": "user", "content": "Explain prefix caching in one sentence."},
14
    ],
15
)
16
print(resp.choices[0].message.content)

Multi-turn

1
messages = [
2
    {"role": "system", "content": "You are a customer-support agent."},
3
    {"role": "user", "content": "I want to change my plan."},
4
    {"role": "assistant", "content": "Sure — which plan are you on right now?"},
5
    {"role": "user", "content": "Standard, and I want to upgrade to Enterprise."},
6
]
7
8
resp = client.chat.completions.create(model="electron", messages=messages)

JSON object output

1
resp = client.chat.completions.create(
2
    model="electron",
3
    messages=[
4
        {"role": "system", "content": "Reply with strict JSON."},
5
        {"role": "user", "content": "List three Indian state capitals as {\"capitals\": [...]}"},
6
    ],
7
    response_format={"type": "json_object"},
8
    temperature=0,
9
)