Chat Completions

POST https://api.shannon-ai.com/v1/chat/completions

Pempho laling'ono kwambiri ndi model id ndi message imodzi ya user.

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.shannon-ai.com/v1",
)

response = client.chat.completions.create(
    model="shannon-3",
    messages=[{"role": "user", "content": "Say hello in one sentence."}],
)

print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://api.shannon-ai.com/v1",
});

const response = await client.chat.completions.create({
  model: "shannon-3",
  messages: [{ role: "user", content: "Say hello in one sentence." }],
});

console.log(response.choices[0].message.content);

curl https://api.shannon-ai.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "shannon-3",
    "messages": [{"role": "user", "content": "Say hello in one sentence."}]
  }'

Yankho ndi JSON object imodzi:

{
  "id": "chatcmpl-5f0c1e7a9b3d4c62a8e1f07d2b46c9a3",
  "object": "chat.completion",
  "created": 1791625200,
  "model": "shannon-3",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello, it is good to meet you.",
        "reasoning_content": "The user wants a greeting in one sentence. Keep it short and friendly."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 1184,
    "completion_tokens": 46,
    "total_tokens": 1230
  }
}

Ma header

Ma header a pempho

Ma header a yankho

Ma field a pempho

messages yokha ndiyofunika. Column ya Imagwira ntchito pa imatchula ma model omwe field imasintha yankho. Hosted open-weight model ndi ma id khumi ndi awiri a mndandanda wa ma model; banja la Shannon 3 ndi shannon-3, shannon-3-pro, shannon-3.1 ndi shannon-3.1-pro. Ma model ndi mitengo

Ma field ena a OpenAI, monga n, user, stream_options, parallel_tool_calls, presence_penalty, frequency_penalty, logit_bias, logprobs, metadata, store ndi prompt_cache_key, amalandiridwa kuti code ya client yomwe ilipo igwire ntchito popanda kusintha. Sasintha yankho: nthawi zonse pamakhala choice imodzi, ndipo stream imatha nthawi zonse ndi usage.

Field yokhala ndi JSON type yolakwika, mwachitsanzo "max_tokens": "100", imabweza 422. Pempho lopanda messages limabwezanso chimodzimodzi.

Ma tool, structured output, reasoning ndi web search chilichonse chili ndi tsamba lake: Kuyitana mafunkseni, Zotuluka zokonzedwa, Reasoning effort, Kusaka pa web.

Pempho lokhala ndi options

Pempho ili limakhazikitsa system message, ma sampling field ndi reasoning effort. Limagwiritsa ntchito hosted open-weight model, yomwe imagwiritsa ntchito zonsezo.

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.shannon-ai.com/v1",
)

response = client.chat.completions.create(
    model="DeepSeek-V4-Flash-0731-W4A16-AUTOROUND-REAP",
    messages=[
        {"role": "system", "content": "You are a physics teacher. Answer in two sentences."},
        {"role": "user", "content": "Why is the sky blue?"},
    ],
    max_tokens=512,
    temperature=0.3,
    top_p=0.9,
    seed=7,
    stop=["\n\n"],
    reasoning_effort="low",
)

message = response.choices[0].message
print(message.reasoning_content)  # the reasoning
print(message.content)            # the answer
print(response.usage)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://api.shannon-ai.com/v1",
});

const response = await client.chat.completions.create({
  model: "DeepSeek-V4-Flash-0731-W4A16-AUTOROUND-REAP",
  messages: [
    { role: "system", content: "You are a physics teacher. Answer in two sentences." },
    { role: "user", content: "Why is the sky blue?" },
  ],
  max_tokens: 512,
  temperature: 0.3,
  top_p: 0.9,
  seed: 7,
  stop: ["\n\n"],
  reasoning_effort: "low",
});

const message = response.choices[0].message;
console.log(message.reasoning_content); // the reasoning
console.log(message.content);           // the answer
console.log(response.usage);

curl https://api.shannon-ai.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "DeepSeek-V4-Flash-0731-W4A16-AUTOROUND-REAP",
    "messages": [
      {"role": "system", "content": "You are a physics teacher. Answer in two sentences."},
      {"role": "user", "content": "Why is the sky blue?"}
    ],
    "max_tokens": 512,
    "temperature": 0.3,
    "top_p": 0.9,
    "seed": 7,
    "stop": ["\n\n"],
    "reasoning_effort": "low"
  }'

Yankho lili ndi mawonekedwe ofanana ndi pamwambapa. usage yake imawonjezera zina ziwiri pa hosted open-weight model: ma prompt token owerengedwa kuchokera ku cache ndi ma token ogwiritsidwa ntchito pa reasoning.

{
  "usage": {
    "prompt_tokens": 31,
    "completion_tokens": 62,
    "total_tokens": 93,
    "prompt_tokens_details": {
      "cached_tokens": 0
    },
    "completion_tokens_details": {
      "reasoning_tokens": 21
    }
  }
}

Kutalika kwa output

max_tokens imachita zinthu ziwiri. Choyamba, ndi chiwerengero cha ma token chomwe chimasungidwa pa balance yanu pempho likayamba. Yankho likatha, chiwerengerocho chimasinthidwa ndi ma token omwe pempho lagwiritsa ntchito. Ngati max_tokens ndi yayikulu kuposa zomwe zatsala pa balance yanu, pempho limabweza 429 Quota exceeded ngakhale yankho lokha likanakwana. Tumizani max_tokens yochepa kuti musunge zochepa.

shannon-coder-1 imawerengedwa mosiyana pa endpoint iyi: pempho lililonse ndi limodzi mwa ma call a Shannon Coder a plan yanu, ndipo palibe ma token omwe amasungidwa. Malire ndi balance

Chachiwiri, imalepheretsa kutalika kwa yankho pa ma model awa:

Popanda max_tokens kapena max_completion_tokens, mtengo ndi 4,096. Pa shannon-coder-1 ndi 65,536.

Messages

Message iliyonse ndi object yokhala ndi role ndi content. content ndi string, kapena array ya ma part pamene message ili ndi zoposa text.

Ndi id ya banja la Shannon 3, ikani malangizo omwe ayenera kutsatiridwa mu message ya user.

Pa ma tier a Shannon, pempho lopanda text ya user komanso lopanda tools limabweza 400 No user message provided.

Ma content part

Kukula, malire ndi mndandanda wathunthu wa mitundu zili patsamba lake. Zithunzi ndi ma file

Object ya yankho

Zifukwa zotsirizira

Yankho losachita stream limanena stop kapena tool_calls.

Usage

Pa hosted open-weight model, prompt_tokens ndi ma message anu ndi matanthauzo a ma tool owerengedwa ndi tokenizer ya model yokha, kuphatikiza ma token a zithunzi zilizonse. Ma endpoint owerengera ma token amabweza nambala yomweyo musanatumize. Kuwerenga ma token

Pa ma tier a Shannon, prompt_tokens imawerenga chilichonse chomwe model idawerenga kuti ilembe yankho, kotero ndi yayikulu kuposa text ya ma message anu yokha.

Streaming

Ndi stream yokhazikitsidwa kukhala true yankho limafika ngati ma event a chat.completion.chunk ndipo limatha ndi data: [DONE]. Chunk yomaliza isanafike imanyamula finish_reason ndi usage; palibe stream_options yofunika. Mawonekedwe a ma chunk, mizere ya keep-alive ndi zolakwika mkati mwa stream zili patsamba lawo. Kutsitsa nthawi yeniyeni

Zolakwika

Cholakwika ndi JSON object yokhala ndi member ya error. Kuyang'ana kumachitika motere: API key, body ya pempho, model id, kenako balance. Table imalemba zomwe endpoint iyi imabweza kawirikawiri. Mndandanda wonse, ndi zoyenera kuyesanso, uli patsamba lake. Zolakwika

{
  "error": {
    "type": "invalid_request_error",
    "message": "unknown model: no-such-model"
  }
}