Qalo e potlakileng

Etsa kopo ea pele: fumana senotlolo sa hau, romela molaetsa o le mong mme u bale karabo. API e bua mebopeho ea OpenAI le Anthropic, kahoo SDK eo u e sebelisang e sebetsa ka mor'a hore u fetole base URL le senotlolo sa eona.

1. Fumana senotlolo sa hau

Akhaonte e 'ngoe le e 'ngoe e na le senotlolo se le seng sa API. Kena, bula Keys & usage mme u se kopitse. Senotlolo se sebetsa ho endpoint e 'ngoe le e 'ngoe le model e 'ngoe le e 'ngoe. Keys & usage

2. Supa client ea hau ho API

Li-endpoint tsohle li ka tlase ho base URL e le 'ngoe. SDK ea OpenAI e e nka e na le /v1 qetellong; SDK ea Anthropic e e nka ntle le eona, hobane e ikentsetsa /v1/messages.

Client	Base URL
Li-SDK tsa OpenAI le li-client tse lumellanang le OpenAI	`https://api.shannon-ai.com/v1`
Li-SDK tsa Anthropic	`https://api.shannon-ai.com`

Li-endpoint tse tharo li hlahisa mongolo. Li fihla ho li-model tse tšoanang; khetha eo sebopeho sa eona se seng se sebelisoang ke code ea hau.

Endpoint	Sebopeho
`POST /v1/chat/completions`	OpenAI Chat Completions
`POST /v1/messages`	Anthropic Messages
`POST /v1/responses`	OpenAI Responses

3. Romela kopo ea pele

Romela molaetsa o le mong oa user ho shannon-3 ho endpoint ea Chat Completions. Nkela YOUR_API_KEY sebaka ka senotlolo sa hau.

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

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": "Hello, Shannon!"}
    ],
    max_tokens=1024
)

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: 'Hello, Shannon!' }
  ],
  max_tokens: 1024
});

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

package main

import (
    "context"
    "fmt"
    openai "github.com/sashabaranov/go-openai"
)

func main() {
    config := openai.DefaultConfig("YOUR_API_KEY")
    config.BaseURL = "https://api.shannon-ai.com/v1"
    client := openai.NewClientWithConfig(config)

    resp, err := client.CreateChatCompletion(
        context.Background(),
        openai.ChatCompletionRequest{
            Model: "shannon-3",
            Messages: []openai.ChatCompletionMessage{
                {Role: "user", Content: "Hello, Shannon!"},
            },
            MaxTokens: 1024,
        },
    )
    if err != nil {
        panic(err)
    }
    fmt.Println(resp.Choices[0].Message.Content)
}

curl -X POST "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": "Hello, Shannon!"}
    ],
    "max_tokens": 1024
  }'

Field	Mofuta	Default	Tlhaloso
`model`	`string`		Id ea model e arabang. Li-id li thathamisitsoe ho Models & pricing.
`messages`	`array`		Moqoqo ho fihlela hajoale: lintho tse nang le `role` le `content`. Molaetsa oa ho qetela oa `user` ke prompt. E hlokahala.
`max_tokens`	`integer`	`4096`	Output budget ea kopo, ho tloha ho 1 ho isa ho 65,536. Kopo e behela palo ena thoko ho balance ea hau mme e lefisoa seo e se sebelisitseng ka 'nete.
`stream`	`boolean`	`false`	`true` e romela karabo sekhechana ka sekhechana ha e ntse e ngoloa.

Body ea kopo ke JSON mme e hloka header Content-Type: application/json. Lenane le felletseng la li-field le leqepheng la endpoint. Chat Completions

Karabo

Kopo e se nang stream e khutlisa ntho e le 'ngoe ea JSON ha karabo e phethahetse. Mengolo le linomoro tse ka tlase ke mehlala.

200 JSON

{
  "id": "chatcmpl-9b2f6c1e4a7d4f0f8a3c5d2e1b6a7c90",
  "object": "chat.completion",
  "created": 1791590400,
  "model": "shannon-3",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello! What would you like to work on?",
        "reasoning_content": "A greeting. A short, friendly reply fits."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 1386,
    "completion_tokens": 74,
    "total_tokens": 1460
  }
}

Field	Mofuta	Tlhaloso
`id`	`string`	`chatcmpl-` e latelwa ke litlhaku tse 32 tsa hexadecimal.
`object`	`string`	Kamehla ke `chat.completion`.
`created`	`integer`	Nako ea karabo ka metsotsoana ea Unix.
`model`	`string`	Id e phatlalalitsoeng ea model e arabileng, esita le ha u rometse alias kapa mongolo o mong.
`choices`	`array`	Kamehla entry e le 'ngoe, e nang le `index` 0.
`choices[0].message.role`	`string`	Kamehla ke `assistant`.
`choices[0].message.content`	`string \| null`	Karabo. E ka ba `null` ha karabo e nka `tool_calls`.
`choices[0].message.reasoning_content`	`string \| null`	Reasoning ea model pele ho karabo ea eona, kapa `null` ha ho e 'ngoe.
`choices[0].message.tool_calls`	`array`	E teng feela ha model e bitsa e 'ngoe ea lisebelisoa tseo u li phatlalalitseng.
`choices[0].finish_reason`	`string`	`stop` bakeng sa karabo, `tool_calls` ha model e bitsa tool.
`usage.prompt_tokens`	`integer`	Tokens tseo model e li balileng bakeng sa karabo ena. Ho `shannon-2-*` le lelapa la Shannon 3 sena se ka feta mongolo oo u o rometseng.
`usage.completion_tokens`	`integer`	Tokens tseo model e li ngotseng.
`usage.total_tokens`	`integer`	Kakaretso. Tefo ea model ea Shannon e baloa ho tloha palong ena.

usage e na le palo ea tokens, hase chelete. Seo kopo e se nkileng ho balance ea hau se bontšoa ka kopo ho Keys & usage; hore na se baloa joang ho leqepheng la Meeli le balance. Meeli le balance
Karabo e 'ngoe le e 'ngoe, esita le phoso le stream, e na le header x-request-id. Haeba kopo ea hau e na le header ea x-request-id, boleng ba eona bo khutla; ho seng joalo boleng ke litlhaku tse 12 tsa hexadecimal. E boloke ho li-log tsa hau: e reha kopo ena e le 'ngoe.

Kopo e tšoanang e le stream

Beha stream ho true ho bala karabo ha e ntse e ngoloa. Tsohle tse ling kopong li lula li tšoana.

from openai import OpenAI

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

stream = client.chat.completions.create(
    model="shannon-3",
    messages=[
        {"role": "user", "content": "Hello, Shannon!"}
    ],
    max_tokens=1024,
    stream=True
)

for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

import OpenAI from 'openai';

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

const stream = await client.chat.completions.create({
  model: 'shannon-3',
  messages: [
    { role: 'user', content: 'Hello, Shannon!' }
  ],
  max_tokens: 1024,
  stream: true
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? '');
}

curl -N -X POST "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": "Hello, Shannon!"}
    ],
    "max_tokens": 1024,
    "stream": true
  }'

Frame e 'ngoe le e 'ngoe ea stream ke mohala o le mong o qalang ka data: mme o na le ntho ea JSON. Frame e le 'ngoe, e gatisitsoeng mela e mengata:

Frame e le 'ngoe

{
  "id": "chatcmpl-9b2f6c1e4a7d4f0f8a3c5d2e1b6a7c90",
  "object": "chat.completion.chunk",
  "created": 1791590400,
  "model": "shannon-3",
  "choices": [
    {
      "index": 0,
      "delta": {
        "content": "Hello! "
      },
      "finish_reason": null
    }
  ]
}

Li-frame, ka tatellano	delta	finish_reason
Ha model e ntse e nahana. Li-frame tse ka bang zero kapa tse ngata.	`{"reasoning_content": "…"}`	`null`
Karabo, sekhechana ka sekhechana. Li-frame tse ka bang zero kapa tse ngata.	`{"content": "…"}`	`null`
Frame ea ho qetela ea JSON. E boetse e nka `usage`; ha u hloke ho e kopa.	`{}`	`"stop"`
Qetello ea stream: mohala oa `data: [DONE]`.

Karabo e na le boemo ba 200 le content-type: text/event-stream. Frame e 'ngoe le e 'ngoe ke mohala data: <json> o latelwang ke mohala o se nang letho.
Mohala oa : ping o fihla ka mor'a metsotsoana e 15 e 'ngoe le e 'ngoe. Ke maikutlo a bulang connection; li-client tsa SSE lia o tlola.
Ho hloleha ka mor'a hore stream e bulehe ho fihla e le frame data: {"error": {…}}. Boemo bo lula bo le 200, kahoo hlahloba li-frame.

Phallela

Haeba kopo e hloleha

Phoso ke ntho ea JSON e nang le error e nang le type le message. Tsena ke likarabo tseo kopo ea pele hangata e kopanang le tsona.

Boemo	Mofuta	Molaetsa	Seo se se bolelang
`401`	`authentication_error`	`Missing authentication`	Ha ho senotlolo se fihlileng ho API. Romela `Authorization: Bearer` e latelwang ke senotlolo sa hau.
`401`	`authentication_error`	`Invalid API key`	Senotlolo hase senotlolo sa hajoale sa akhaonte. Se kopitse hape ho tsoa Keys & usage.
`400`	`invalid_request_error`	`unknown model: <id>`	`model` hase id e phatlalalitsoeng. Sebelisa id ho tsoa Models & pricing.
`422`	`invalid_request_error`	`Failed to deserialize the JSON body into the target type…`	Field e na le mofuta o fosahetseng oa JSON, kapa `messages` ha e eo.
`429`	`rate_limit_error`	`Quota exceeded. Upgrade your plan at shannon-ai.com/plan`	Output budget ea kopo ha e kene ho balance ea hau.

Diphoso

Moo u ka yang teng ka mor'a moo

Li-model le litheko — Khetha model: li-id, context window le litheko.
Chat Completions — Field e 'ngoe le e 'ngoe ea kopo, field ea karabo le event ea stream ea endpoint eo u e bitsitseng.
Phallela — Li-frame tsa stream, mela ea keep-alive le liphoso ka hare ho stream.
Mekhosi ya mesebetsi — Lumella model hore e bitse li-function tsa hau.
Meeli le balance — Seo kopo e se behelang thoko le seo e se jang, le hore na u ka romela likopo tse kae.
API Playground — Leka likopo ka browser ka senotlolo sa hau.