Chat Completions

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

Kopo e nyane ka ho fetisisa ke id ea model le molaetsa o le mong oa 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."}]
  }'

Karabo ke ntho e le 'ngoe ea JSON:

{
  "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
  }
}

Li-header

Li-header tsa kopo

Li-header tsa karabo

Li-field tsa kopo

messages feela e hlokahala. Kholomo ea E sebelisoa ke e reha li-model tseo field e fetolang karabo ho tsona. Li-model tse hosted tsa open-weight ke li-id tse leshome le metso e 'meli tsa lenane la li-model; lelapa la Shannon 3 ke shannon-3, shannon-3-pro, shannon-3.1 le shannon-3.1-pro. Li-model le litheko

Li-field tse ling tsa OpenAI, tse kang n, user, stream_options, parallel_tool_calls, presence_penalty, frequency_penalty, logit_bias, logprobs, metadata, store le prompt_cache_key, lia amohelwa e le hore code ea moreki e teng e sebetse ntle le phetoho. Ha li fetole karabo: kamehla ho na le choice e le 'ngoe, mme stream e lula e fela ka usage.

Field e nang le mofuta o fosahetseng oa JSON, mohlala "max_tokens": "100", e khutlisa 422. Kopo e se nang messages le eona e joalo.

Tools, structured output, reasoning le web search e 'ngoe le e 'ngoe e na le leqephe la eona: Mekhosi ya mesebetsi, Diphumo tse hlophisitsoeng, Effort ea reasoning, Patlo ya webe.

Kopo e nang le likhetho

Kopo ena e beha molaetsa oa system, li-field tsa sampling le reasoning effort. E sebelisa model e hosted ea open-weight, e sebelisang tsohle.

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"
  }'

Karabo e na le sebopeho se tšoanang le se ka holimo. usage ea eona e eketsa lintlha tse peli ho li-model tse hosted tsa open-weight: prompt tokens tse baloang ho cache le tokens tse sebelisitsoeng ho reasoning.

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

Bolelele ba output

max_tokens e etsa lintho tse peli. Ea pele, ke palo ea tokens e behelloang thoko ho balance ea hau ha kopo e qala. Ha karabo e phethahetse, palo eo e nkeloa sebaka ke tokens tseo kopo e li sebelisitseng. Haeba max_tokens e kholo ho feta se setseng ho balance ea hau, kopo e khutlisa 429 Quota exceeded esita le ha karabo ka boeona e ne e tla lekana. Romela max_tokens e tlase ho beha thoko tse fokolang.

shannon-coder-1 e baloa ka tsela e fapaneng endpoint-ng ena: kopo e 'ngoe le e 'ngoe ke e 'ngoe ea likopo tsa Shannon Coder tsa plan ea hau, mme ha ho tokens tse behelloang thoko bakeng sa eona. Meeli le balance

Ea bobeli, e fokotsa bolelele ba karabo ho li-model tsena:

Ntle le max_tokens kapa max_completion_tokens, boleng ke 4,096. Ho shannon-coder-1 ke 65,536.

Melaetsa

Molaetsa o mong le o mong ke ntho e nang le role le content. content ke string, kapa array ea likarolo ha molaetsa o nka ho feta mongolo.

Ka id ea lelapa la Shannon 3, kenya litaelo tse lokelang ho latelloa molaetsa oa user.

Ho li-tier tsa Shannon kopo e se nang mongolo oa user kapa tools e khutlisa 400 No user message provided.

Likarolo tsa litaba

Boholo, meeli le lenane le felletseng la mebopeho li na le leqephe la tsona. Litšoantšo le lifaele

Ntho ea karabo

Mabaka a ho fela

Karabo e se nang streaming e tlaleha stop kapa tool_calls.

Usage

Ho li-model tse hosted tsa open-weight, prompt_tokens ke melaetsa ea hau le litlhaloso tsa tool tse baloang ka tokenizer ea model ka boeona, hammoho le tokens tsa litšoantšo leha e le life. Li-endpoint tsa ho bala tokens li khutlisa palo e tšoanang pele u romela. Ho bala tokens

Ho li-tier tsa Shannon, prompt_tokens e bala tsohle tseo model e li balileng ho ngola karabo, kahoo e kholo ho feta mongolo oa melaetsa ea hau feela.

Streaming

Ha stream e behiloe ho true karabo e fihla e le li-event tsa chat.completion.chunk mme e fela ka data: [DONE]. Chunk ea ho qetela pele ho eona e nka finish_reason le usage; ha ho hlokahale stream_options. Mebopeho ea li-chunk, mela ea keep-alive le liphoso ka hare ho stream li na le leqephe la tsona. Phallela

Liphoso

Phoso ke ntho ea JSON e nang le setho sa error. Liphatlalatso li etsoa ka tatellano ena: senotlolo sa API, body ea kopo, id ea model, ebe balance. Tafole e thathamisa seo endpoint ena e se khutlisang hangata. Lenane le felletseng, le se ka lekoang hape, le na le leqephe la lona. Diphoso

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