Chat Completions

Generate model responses from a list of messages. The core Tolka Edge endpoint, fully OpenAI-compatible with streaming support.

Create a model response for a conversation using your dedicated AI Rig. This is Tolka Edge's primary endpoint and mirrors the OpenAI Chat Completions contract exactly.

POSThttps://api.tolkaedge.com/v1/chat/completions

Headers

Header	Required	Description
`Authorization`	Yes	`Bearer tk-live-...`
`Content-Type`	Yes	`application/json`

Request body

Parameter	Type	Description
`model`required	string	Model slug to route to. Currently supported: `Qwen/Qwen3-14B`.
`messages`required	array	The conversation so far. Each item has a role (system, user, assistant) and content.
`stream`	boolean	Stream partial deltas as server-sent events.Default: `false`
`temperature`	number	Sampling temperature between 0 and 2. Higher is more random.Default: `1`
`max_tokens`	integer	Maximum number of tokens to generate in the completion.

Tolka Edge forwards standard OpenAI fields directly to the vLLM instance running on your AI Rig. Advanced payload fields are passed through verbatim.

Message object

Parameter	Type	Description
`role`required	string	One of system, user, or assistant.
`content`required	string	The message text.

Request example

from openai import OpenAI
import os
 
client = OpenAI(
    base_url="https://api.tolkaedge.com/v1",
    api_key=os.environ.get("TOLKA_API_KEY"),
)
 
resp = client.chat.completions.create(
    model="Qwen/Qwen3-14B",
    messages=[
        {"role": "system", "content": "You are a concise assistant."},
        {"role": "user", "content": "Explain dedicated GPU inference."},
    ],
    temperature=0.7,
    max_tokens=256,
)
 
print(resp.choices[0].message.content)

import OpenAI from "openai";
 
const client = new OpenAI({ baseURL: "https://api.tolkaedge.com/v1", apiKey: process.env.TOLKA_API_KEY });
 
const resp = await client.chat.completions.create({
  model: "Qwen/Qwen3-14B",
  messages: [
    { role: "system", content: "You are a concise assistant." },
    { role: "user", content: "Explain dedicated GPU inference." },
  ],
  temperature: 0.7,
  max_tokens: 256,
});
console.log(resp.choices[0].message.content);

curl https://api.tolkaedge.com/v1/chat/completions \
  -H "Authorization: Bearer $TOLKA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Qwen/Qwen3-14B",
    "messages": [
      {"role": "system", "content": "You are a concise assistant."},
      {"role": "user", "content": "Explain dedicated GPU inference."}
    ],
    "temperature": 0.7,
    "max_tokens": 256
  }'

Response

{
  "id": "chatcmpl-9f2b...",
  "object": "chat.completion",
  "created": 1716492000,
  "model": "Qwen/Qwen3-14B",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Dedicated GPU inference means running your AI models on isolated hardware, ensuring you don't share compute or memory with other users, which guarantees privacy and consistent latency."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 14,
    "completion_tokens": 30,
    "total_tokens": 44
  }
}

Tolka Edge Headers

Every response includes custom headers for cost and debugging:

Header	Example	Description
`X-Tolka-Cost-INR`	`0.0000`	The exact cost billed to your wallet in paise.
`X-Tolka-Model-Used`	`Qwen/Qwen3-14B`	The model slug that was used.
`X-Tolka-Provider-Used`	`runpod`	The cloud provider hosting your dedicated AI Rig.