Chat Completions

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

Ny fangatahana kely indrindra dia model id sy hafatra user iray.

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."}]
  }'

Ny valiny dia objet JSON iray:

{
  "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
  }
}

Headers

Headers amin'ny fangatahana

Headers amin'ny valiny

Sahan'ny fangatahana

Ny messages ihany no ilaina. Ny tsanganana Ampiharin'ny dia manonona ny model izay anovan'ny saha ny valiny. Ny model open-weight hosted dia ny id roa ambin'ny folo ao amin'ny lisitry ny model; ny fianakavian'i Shannon 3 dia shannon-3, shannon-3-pro, shannon-3.1 ary shannon-3.1-pro. Model & vidiny

Ny saha OpenAI hafa, toy ny n, user, stream_options, parallel_tool_calls, presence_penalty, frequency_penalty, logit_bias, logprobs, metadata, store ary prompt_cache_key, dia ekena mba hiasa tsy miova ny kaody client efa misy. Tsy manova ny valiny izy ireo: choice iray foana no misy, ary ny stream dia mifarana foana miaraka amin'ny usage.

Ny saha manana karazana JSON diso, ohatra "max_tokens": "100", dia mamerina 422. Toy izany koa ny fangatahana tsy misy messages.

Ny tools, structured output, reasoning ary web search dia samy manana pejy manokana: Antso fiasa, Vokatra voarafitra, Reasoning effort, Fikarohana web.

Fangatahana misy options

Ity fangatahana ity dia mametra hafatra system, ny saha sampling ary ny reasoning effort. Mampiasa model open-weight hosted izy, izay mampihatra azy rehetra.

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"
  }'

Ny valiny dia manana endrika mitovy amin'ny etsy ambony. Ny usage-ny dia manampy antsipiriany roa amin'ny model open-weight hosted: ny prompt tokens novakina avy ao amin'ny cache sy ny tokens lany tamin'ny reasoning.

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

Halavan'ny output

Ny max_tokens dia manao zavatra roa. Voalohany, izy no isan'ny tokens tazonina avy amin'ny balance-nao rehefa manomboka ny fangatahana. Rehefa vita ny valiny, io habetsaka io dia soloina ny tokens nampiasain'ny fangatahana. Raha lehibe noho ny sisa amin'ny balance-nao ny max_tokens, ny fangatahana dia mamerina 429 Quota exceeded na dia ho tafiditra aza ny valiny. Alefaso max_tokens ambany kokoa mba hitazonana tsy dia betsaka.

Ny shannon-coder-1 dia isaina amin'ny fomba hafa amin'ity endpoint ity: ny fangatahana tsirairay dia iray amin'ny antso Shannon Coder an'ny plan-nao, ary tsy misy tokens tazonina ho azy. Fetra sy balance

Faharoa, mametra ny halavan'ny valiny amin'ireto model ireto izy:

Raha tsy misy max_tokens na max_completion_tokens, 4,096 ny sanda. Amin'ny shannon-coder-1 dia 65,536.

Hafatra

Ny hafatra tsirairay dia objet misy role sy content. Ny content dia string, na array misy ampahany rehefa mitondra zavatra mihoatra ny soratra ny hafatra.

Amin'ny id ao amin'ny fianakavian'i Shannon 3, apetraho ao amin'ny hafatra user ny toromarika tsy maintsy arahina.

Amin'ny tier Shannon dia mamerina 400 No user message provided ny fangatahana tsy misy soratra user sy tsy misy tools.

Content parts

Ny habe, ny fetra ary ny lisitra feno amin'ny endrika dia manana pejy manokana. Sary sy rakitra

Ny objet valiny

Antony fifaranana

Ny valiny tsy streamed dia milaza stop na tool_calls.

Usage

Amin'ny model open-weight hosted, ny prompt_tokens dia ny hafatra sy famaritana tool nataonao nisaina tamin'ny tokenizer an'ny model, ampiana ny tokens an'ny sary rehetra. Ny endpoint fanisana tokens dia mamerina isa mitovy alohan'ny handefasanao. Fanisana token

Amin'ny tier Shannon, ny prompt_tokens dia manisa izay rehetra novakin'ny model hanoratana ny valiny, ka lehibe noho ny soratry ny hafatra nataonao ihany izy.

Streaming

Rehefa apetraka true ny stream dia tonga ho events chat.completion.chunk ny valiny ary mifarana amin'ny data: [DONE]. Ny chunk farany alohany dia mitondra finish_reason sy usage; tsy ilaina ny stream_options. Ny endrika chunk, ny andalana keep-alive ary ny hadisoana ao anaty stream dia manana pejy manokana. Fandefasana mivantana

Hadisoana

Ny hadisoana dia objet JSON misy mpikambana error. Mandeha araka ity filaharana ity ny fanamarinana: API key, vatan'ny fangatahana, model id, avy eo balance. Ny tabilao dia mampiseho izay averin'ity endpoint ity matetika indrindra. Ny lisitra feno, miaraka amin'izay tokony hoeranina, dia manana pejy manokana. Fitantanana hadisoana

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