Responses API

POST https://api.shannon-ai.com/v1/responses

L-iżgħar talba hija id ta' mudell u test. It-tweġiba ma tiġix fi stream sakemm ma titlobx.

from openai import OpenAI

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

response = client.responses.create(
    model="shannon-3",
    input="Say hello in one sentence.",
)

print(response.output_text)

import OpenAI from "openai";

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

const response = await client.responses.create({
  model: "shannon-3",
  input: "Say hello in one sentence.",
});

console.log(response.output_text);

curl https://api.shannon-ai.com/v1/responses \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "shannon-3",
    "input": "Say hello in one sentence."
  }'

It-tweġiba hija oġġett tat-tweġiba. L-output tagħha jżomm ir-raġunament, meta l-mudell irraġuna, u l-messaġġ:

{
  "id": "resp_5f0c1e7a9b3d4c62a8e1f07d2b46c9a3",
  "object": "response",
  "created_at": 1791625200,
  "model": "shannon-3",
  "status": "completed",
  "error": null,
  "incomplete_details": null,
  "instructions": null,
  "metadata": {},
  "parallel_tool_calls": true,
  "temperature": null,
  "tool_choice": "auto",
  "tools": [],
  "top_p": null,
  "output": [
    {
      "id": "rs_8a2e6c1f4b7d4e03b9c5a1d70f3e2b64",
      "type": "reasoning",
      "summary": [
        {
          "type": "summary_text",
          "text": "The user wants a greeting in one sentence. Keep it short and friendly."
        }
      ]
    },
    {
      "id": "msg_1c7b9e3a5d2f4a86b0e4c8d19a6f7e25",
      "type": "message",
      "role": "assistant",
      "status": "completed",
      "content": [
        {
          "type": "output_text",
          "text": "Hello, it is good to meet you.",
          "annotations": []
        }
      ]
    }
  ],
  "usage": {
    "input_tokens": 1184,
    "output_tokens": 46,
    "total_tokens": 1230
  }
}

Kampi tat-talba

model u input huma dak li teħtieġ talba. Il-kolonna Japplikah issemmi l-mudelli li fuqhom qasam jibdel it-tweġiba. Il-mudelli open-weight ospitati huma t-tnax-il id tal-lista tal-mudelli; il-familja Shannon 3 hija shannon-3, shannon-3-pro, shannon-3.1 u shannon-3.1-pro. Mudelli u prezzijiet

previous_response_id, store, metadata, parallel_tool_calls, truncation, include, user u prompt_cache_key jiġu aċċettati biex il-kodiċi eżistenti tal-klijent jaħdem mingħajr tibdil. Ma jibdlux it-tweġiba. Peress li ma tinħażen l-ebda konversazzjoni, talba ta' segwitu ġġorr it-turns preċedenti f'input.

Input items

Input item huwa jew messaġġ, b'role u content, jew element tipizzat li jġorr sejħa ta' tool jew ir-riżultat tagħha. content huwa string jew array ta' partijiet.

Poġġi l-istruzzjonijiet fil-qasam instructions. B'id tal-familja Shannon 3, poġġi l-istruzzjonijiet li jridu jiġu rispettati fil-messaġġ user.

L-elementi ta' output ta' tweġiba jistgħu jmorru lura f'input kif inhuma: l-elementi message u function_call jinqraw, l-elementi reasoning jitqabżu.

Partijiet tal-kontenut

Fuq il-livelli Shannon l-immaġini u l-fajls jinqraw mill-aħħar element user ta' input. Id-daqsijiet u l-limiti għandhom paġna tagħhom. Immaġini u fajls

Talba b'struzzjonijiet u għażliet

Din it-talba tibgħat struzzjonijiet, konversazzjoni bħala input items, limitu tal-output, temperatura u effort tar-raġunament. Tuża mudell open-weight ospitat, li japplika kollha.

from openai import OpenAI

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

response = client.responses.create(
    model="DeepSeek-V4-Flash-0731-W4A16-AUTOROUND-REAP",
    instructions="You are a physics teacher. Answer in two sentences.",
    input=[
        {"role": "user", "content": "Why is the sky blue?"},
        {"role": "assistant", "content": "Air scatters blue light more than red light."},
        {"role": "user", "content": "And why are sunsets red?"},
    ],
    max_output_tokens=512,
    temperature=0.3,
    reasoning={"effort": "low"},
)

for item in response.output:
    if item.type == "reasoning":
        print("reasoning:", item.summary[0].text)
    elif item.type == "message":
        print("answer:", item.content[0].text)
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.responses.create({
  model: "DeepSeek-V4-Flash-0731-W4A16-AUTOROUND-REAP",
  instructions: "You are a physics teacher. Answer in two sentences.",
  input: [
    { role: "user", content: "Why is the sky blue?" },
    { role: "assistant", content: "Air scatters blue light more than red light." },
    { role: "user", content: "And why are sunsets red?" },
  ],
  max_output_tokens: 512,
  temperature: 0.3,
  reasoning: { effort: "low" },
});

for (const item of response.output) {
  if (item.type === "reasoning") console.log("reasoning:", item.summary[0].text);
  if (item.type === "message") console.log("answer:", item.content[0].text);
}
console.log(response.usage);

curl https://api.shannon-ai.com/v1/responses \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "DeepSeek-V4-Flash-0731-W4A16-AUTOROUND-REAP",
    "instructions": "You are a physics teacher. Answer in two sentences.",
    "input": [
      {"role": "user", "content": "Why is the sky blue?"},
      {"role": "assistant", "content": "Air scatters blue light more than red light."},
      {"role": "user", "content": "And why are sunsets red?"}
    ],
    "max_output_tokens": 512,
    "temperature": 0.3,
    "reasoning": {"effort": "low"}
  }'

It-tweġiba ta' mudell open-weight ospitat għandha id, object, model, created_at, status, output u usage. L-usage tagħha żżid it-tokens moqrija mill-cache u t-tokens użati għar-raġunament:

{
  "usage": {
    "input_tokens": 58,
    "input_tokens_details": {
      "cached_tokens": 0
    },
    "output_tokens": 67,
    "output_tokens_details": {
      "reasoning_tokens": 19
    },
    "total_tokens": 125
  }
}

L-oġġett tat-tweġiba

id, object, status, output u usage jinsabu f'kull tweġiba. Ibni l-kodiċi tiegħek fuqhom.

Output items

L-SDKs ta' OpenAI jgħaqqdu t-test tal-messaġġ għalik f'response.output_text.

Streaming

Issettja stream għal true għal server-sent events. Kull frame għandu linja event: bl-isem tal-avveniment u linja data: b'oġġett JSON; l-oġġett jirrepeti l-isem f'type. Qabbel id-deltas mal-element tagħhom b'output_index.

Mal-SDK ta' OpenAI, aqra l-avvenimenti skont type tagħhom:

from openai import OpenAI

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

stream = client.responses.create(
    model="shannon-3",
    input="Explain server-sent events in three sentences.",
    stream=True,
)

for event in stream:
    kind = getattr(event, "type", "error")     # an event without a type is an error event
    if kind == "response.reasoning_summary_text.delta":
        pass                                   # a piece of reasoning: event.delta
    elif kind == "response.output_text.delta":
        print(event.delta, end="", flush=True)
    elif kind == "response.output_item.done" and event.item.type == "function_call":
        print("\ntool call:", event.item.name, event.item.arguments)
    elif kind == "response.completed":
        print("\n", event.response.usage)
    elif kind in ("response.failed", "error"):
        raise RuntimeError("the stream failed, send the request again")

import OpenAI from "openai";

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

const stream = await client.responses.create({
  model: "shannon-3",
  input: "Explain server-sent events in three sentences.",
  stream: true,
});

for await (const event of stream) {
  if (event.type === "response.reasoning_summary_text.delta") {
    // a piece of reasoning: event.delta
  } else if (event.type === "response.output_text.delta") {
    process.stdout.write(event.delta);
  } else if (event.type === "response.output_item.done" && event.item.type === "function_call") {
    console.log("\ntool call:", event.item.name, event.item.arguments);
  } else if (event.type === "response.completed") {
    console.log("\n", event.response.usage);
  } else if (event.type === "response.failed" || event.type === "error") {
    throw new Error("the stream failed, send the request again");
  }
}

curl -N https://api.shannon-ai.com/v1/responses \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "shannon-3",
    "input": "Explain server-sent events in three sentences.",
    "stream": true
  }'

L-avvenimenti jaslu f'dan l-ordni. Il-gruppi għar-raġunament, il-messaġġ u s-sejħiet ta' tools jidhru biss meta t-tweġiba jkollha.

event: response.created
data: {"type":"response.created","response":{"id":"resp_5f0c…","object":"response","status":"in_progress","output":[], …}}

event: response.in_progress
data: {"type":"response.in_progress","response":{"id":"resp_5f0c…","object":"response","status":"in_progress","output":[], …}}

event: response.output_item.added
data: {"type":"response.output_item.added","output_index":0,"item":{"id":"msg_1c7b…","type":"message","role":"assistant","status":"in_progress","content":[]}}

event: response.content_part.added
data: {"type":"response.content_part.added","output_index":0,"content_index":0,"part":{"type":"output_text","text":""}}

event: response.output_text.delta
data: {"type":"response.output_text.delta","output_index":0,"content_index":0,"delta":"Server-sent"}

event: response.output_text.delta
data: {"type":"response.output_text.delta","output_index":0,"content_index":0,"delta":" events let a server push text over one HTTP reply."}

event: response.output_text.done
data: {"type":"response.output_text.done","output_index":0,"content_index":0,"text":"Server-sent events let a server push text over one HTTP reply."}

event: response.content_part.done
data: {"type":"response.content_part.done","output_index":0,"content_index":0,"part":{"type":"output_text","text":"Server-sent events let a server push text over one HTTP reply."}}

event: response.output_item.done
data: {"type":"response.output_item.done","output_index":0,"item":{"id":"msg_1c7b…","type":"message","role":"assistant","status":"completed","content":[{"type":"output_text","text":"Server-sent events let a server push text over one HTTP reply.","annotations":[]}]}}

event: response.completed
data: {"type":"response.completed","response":{"id":"resp_5f0c…","object":"response","status":"completed","output":[…],"usage":{"input_tokens":1190,"output_tokens":18,"total_tokens":1208}, …}}

Stream li jfalli

Fuq il-livelli Shannon tweġiba fallita tispiċċa b'response.failed minflok response.completed. L-response.status tagħha huwa failed u response.error iżomm code, rate_limit_exceeded jew server_error, u message.

Fuq il-mudelli open-weight ospitati falliment huwa avveniment error b'code api_error u message. response.completed xorta jiġi warajh.

Qis stream bħala fallit meta fih avveniment response.failed jew error, u ibgħat it-talba mill-ġdid.

Tools

Tool hija funzjoni li l-kodiċi tiegħek iħaddem. Iddeskriviha f'tools; meta l-mudell jeħtieġha, it-tweġiba żżomm element function_call minflok, jew maġenb, il-messaġġ.

Definizzjoni ta' tool f'tools:

[
  {
    "type": "function",
    "name": "get_weather",
    "description": "Current weather for a city.",
    "parameters": {
      "type": "object",
      "properties": {
        "city": {
          "type": "string"
        }
      },
      "required": [
        "city"
      ]
    }
  }
]

L-output ta' tweġiba li tsejjaħha:

[
  {
    "id": "call_3d9a7c1e5b2f4a60c8e1d7f09b24a6c5",
    "type": "function_call",
    "name": "get_weather",
    "call_id": "call_3d9a7c1e5b2f4a60c8e1d7f09b24a6c5",
    "arguments": "{\"city\":\"Paris\"}"
  }
]

Ħaddem il-funzjoni, imbagħad ibgħat it-talba mill-ġdid bis-sejħa u r-riżultat tagħha miżjuda f'input:

[
  {
    "role": "user",
    "content": "What is the weather in Paris?"
  },
  {
    "type": "function_call",
    "call_id": "call_3d9a7c1e5b2f4a60c8e1d7f09b24a6c5",
    "name": "get_weather",
    "arguments": "{\"city\":\"Paris\"}"
  },
  {
    "type": "function_call_output",
    "call_id": "call_3d9a7c1e5b2f4a60c8e1d7f09b24a6c5",
    "output": "{\"temperature_c\": 18, \"sky\": \"cloudy\"}"
  }
]

Jinqraw biss tools tat-tip function. Fuq shannon-1.6-*, shannon-2-* u shannon-coder-1 l-iskema parameters titnaqqas għall-keywords komuni ta' JSON Schema: oneOf jinqara bħala anyOf, u keywords bħal $ref, additionalProperties, const, default u title ma jiġux mgħoddija lill-mudell.

Il-valuri required u funzjoni msemmija ta' tool_choice jiġu applikati mill-mudelli open-weight ospitati.

Iċ-ċiklu tat-tools huwa deskritt kollu fil-paġna tiegħu. Sejħiet ta' Funzjonijiet

Output strutturat

Dan il-format jikteb l-output strutturat bħala text.format. L-iskema tinsab direttament f'format, maġenb name u strict, mhux taħt membru json_schema bħal fuq Chat Completions.

• {"type": "json_object"} jitlob kwalunkwe oġġett JSON.
• {"type": "json_schema", "name", "schema", "strict"} jitlob JSON li jsegwi l-iskema tiegħek. strict huwa true meta ma tibgħatux.

{
  "model": "shannon-1.6-lite",
  "input": "Extract the city and the date: 'The conference opens in Lisbon on 12 May 2027.'",
  "text": {
    "format": {
      "type": "json_schema",
      "name": "event",
      "strict": true,
      "schema": {
        "type": "object",
        "properties": {
          "city": {
            "type": "string"
          },
          "date": {
            "type": "string",
            "description": "ISO 8601 date"
          }
        },
        "required": [
          "city",
          "date"
        ]
      }
    }
  }
}

Il-JSON jasal bħala test fil-parti output_text tal-messaġġ; parsjah int. Mudell open-weight ospitat mingħajr output strutturat jagħti lura 400.

{"city": "Lisbon", "date": "2027-05-12"}

Liema mudell jaċċetta liema forma hija elenkata fil-paġna tagħha. Outputs Strutturati

Raġunament

Mudell li jirraġuna jagħti lura r-raġunament tiegħu bħala output item tat-tip reasoning, qabel il-messaġġ. Fi stream jasal bħala avvenimenti response.reasoning_summary_text.delta.

Il-mudelli open-weight ospitati japplikaw reasoning.effort: off, low, medium jew high, b'high meta ma tibgħat xejn. off ma jagħti lura l-ebda element ta' raġunament. Ibgħatu bħala oġġett; string waħdu ma jinqarax fuq dan l-endpoint.

L-efforts huma deskritti fil-paġna tagħhom. Sforz ta' raġunament

Limitu tal-output

max_output_tokens huwa 4,096 meta ma tibgħatux. Huwa l-għadd ta' tokens irriżervat mill-bilanċ tiegħek meta tibda t-talba; meta t-tweġiba tkun kompluta, dak l-ammont jiġi mibdul b'dak li użat it-talba. Valur li ma jidholx fil-bilanċ li fadal tiegħek jagħti lura 429 Quota exceeded.

Jillimita t-tul tat-tweġiba fuq shannon-1.6-*, shannon-2-* u shannon-coder-1, u t-tul tat-test tat-tweġiba fuq il-mudelli open-weight ospitati.

Fuq dan l-endpoint shannon-coder-1 jitħallas f'tokens mill-bilanċ tiegħek, bħal kull mudell ieħor.

Body ta' talba jista' jkun sa 32 MiB. Limiti u bilanċ

Żbalji

L-iżbalji jużaw il-body tal-iżball ta' OpenAI: oġġett error b'type u message; xi wħud iġorru wkoll code u param bil-valur null. It-tabella telenka dak li dan l-endpoint jagħti lura l-aktar spiss. Il-lista kompluta, b'dak li għandu jerġa' jiġi pprovat, għandha paġna tagħha. Ġestjoni tal-iżbalji