Gafee gaa n'ọdịnaya
Images na files

Images na files

Zipụ images na ozi: models na-anara ha, ụdị anabatara na limits ha.

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

Image na-aga n'ime user message, dị ka otu akụkụ nke content ya. Sample na-agụ file dị na kọmputa ma zipụ ya dị ka data URL.

import base64
from openai import OpenAI

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

with open("photo.jpg", "rb") as f:
    image = base64.b64encode(f.read()).decode()

response = client.chat.completions.create(
    model="shannon-3",
    messages=[{
        "role": "user",
        "content": [
            {"type": "text", "text": "What is in this picture?"},
            {"type": "image_url", "image_url": {"url": "data:image/jpeg;base64," + image}},
        ],
    }],
)

print(response.choices[0].message.content)
200 JSON
{
  "id": "chatcmpl-8d1f3a5c7e9b4d2f6a8c0e2b4d6f8a1c",
  "object": "chat.completion",
  "created": 1791590400,
  "model": "shannon-3",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "A grey cat asleep on a windowsill, with a potted fern beside it.",
        "reasoning_content": null
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 286,
    "completion_tokens": 19,
    "total_tokens": 305
  }
}

Models na-anara images

Models Na-agụ
shannon-3, shannon-3-pro, shannon-3.1, shannon-3.1-pro Images, na akwụkwọ: PDF, Word, PowerPoint, Excel na plain-text files.
shannon-1.6-lite, shannon-1.6-pro Images.
shannon-2-lite, shannon-2-pro Images, na requests na-ezipukwa tools ma ọ bụ response_format.
Kimi-K3-3BIT-REAP, MiniMax-M3-3BIT-REAP, Kimi-K2.6-W4A16-AUTOROUND-REAP, inkling-W4A16-AUTOROUND-REAP, MiMo-V2.5-W8A16 Images.
DeepSeek-V4-Pro-0813-3BIT-REAP, GLM-5.2-3BIT-REAP, Nemotron3Ultra-3BIT-REAP, DeepSeek-V4-Flash-0731-W4A16-AUTOROUND-REAP, Laguna-S-2.1-W4A16-AUTOROUND-REAP, MiMo-V2.5-Pro-W8A16, Hy3-W8A16 Text. A na-aza request nwere image na status 400.
shannon-coder-1 Text.

GET /v1/models na-akọ image input maka id ọ bụla dị ka capabilities.vision. Models & ọnụahịa

  • Shannon models na-agụ files nke user message ikpeazụ. Tinye image ahụ n'ozi na-ajụ banyere ya.
  • Shannon 3 family na-edebekwa ole na ole images na akwụkwọ nke user messages gara aga n'anya mgbe e zipụrụ ha inline, ma na-agụ oke ọnụọgụ images maka request, a na-agụ site na nke mbụ.
  • Hosted open-weight models na-agụ images nke ozi ọ bụla nke mkparịta ụka ahụ.

Ụdị anabatara

A na-ezipu file n'otu n'ụzọ abụọ: n'ime request dị ka base64, ma ọ bụ dị ka adreesị API na-ebubata. File ọ bụla na-eso request; API enweghị upload endpoint.

Endpoint N'ime request (base64) Site na adreesị
/v1/chat/completions {"type": "image_url", "image_url": {"url": "data:image/jpeg;base64,…"}} {"type": "image_url", "image_url": {"url": "https://…"}}
/v1/responses {"type": "input_image", "image_url": "data:image/jpeg;base64,…"} {"type": "input_image", "image_url": "https://…"}
/v1/messages {"type": "image", "source": {"type": "base64", "media_type": "image/jpeg", "data": "…"}} {"type": "image", "source": {"type": "url", "url": "https://…"}}

Otu user message ahụ n'format ọ bụla:

{
  "role": "user",
  "content": [
    {"type": "text", "text": "What is in this picture?"},
    {
      "type": "image_url",
      "image_url": {
        "url": "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD…"
      }
    }
  ]
}
  • Data URL nwere ụdị data:<media type>;base64,<data>. Akụkụ ;base64 achọrọ.
  • Na /v1/chat/completions na /v1/responses, image_url nwere ike ịbụ string n'onwe ya ma ọ bụ object {"url": "…"}.
  • File type si na data URL ma ọ bụ site na media_type. Na-enweghị otu, a na-agụ image dị ka image/png.
  • Inline files na-agụ n'ime nha nke request body, nke nwere ike ịbụ ruo 32 MiB.
  • Anabatara maka ndakọrịta: image_url.detail.

Limits maka remote files

Mgbe ị zipụrụ adreesị, API na-ebubata file tupu model agba, n'ime limits ndị a:

Oke Iwu
Nha Ruo 8 MiB maka file ọ bụla.
Oge Sekọnd 20 iji aza.
Redirects Ihe na-akarịghị 5.
Adreesị http:// ma ọ bụ https://, na-enweghị user name ma ọ bụ paswọọdụ n'adreesị, na host nwere adreesị ọha.
Azịza nke host Status n'ọkwa 200 na body na-abụghị efu. A na-agụ file type site na Content-Type.
A na-arịọ dị ka User-Agent: ShannonBot/1.0 (+https://shannon-ai.com)

Otu images si agụ na tokens

Na hosted open-weight models, image na-agụ otu token maka patch pixel 28 × 28 ọ bụla: obosara kewara 28, gbagoro elu, ugboro ogologo kewara 28, gbagoro elu. Result bụ akụkụ nke usage.prompt_tokens a na-akwụ ụgwọ ya dị ka input.

1024 × 1024
1,369 tokens
1920 × 1080
2,691 tokens
512 × 512
361 tokens
  • A na-ebubata remote image buru ụzọ, ya mere a na-agụ nha ya n'ezie.
  • Image enweghị ike ịgụ nha ya na-agụ tokens 1,024.
  • Counting endpoints na-eji otu iwu ahụ maka data URLs. Ha anaghị ebubata remote address ma na-agụ ya dị ka tokens 1,024. Ịgụ tokens
  • Maka Shannon model, gụọ ihe request nwere images tụ n'ime usage n'azịza ahụ.

Akwụkwọ na files ndị ọzọ

Shannon 3 family (shannon-3, shannon-3-pro, shannon-3.1, shannon-3.1-pro) na-agụkwa akwụkwọ. A na-ewepụta ederede nke akwụkwọ ahụ nye model ya na ozi gị.

Akwụkwọ Media type
PDF application/pdf
Word application/vnd.openxmlformats-officedocument.wordprocessingml.document
PowerPoint application/vnd.openxmlformats-officedocument.presentationml.presentation
Excel application/vnd.openxmlformats-officedocument.spreadsheetml.sheet

Akwụkwọ bụ akụkụ nke user message, dị ka image:

Endpoint N'ime request (base64) Site na adreesị
/v1/chat/completions, /v1/messages {"type": "document", "source": {"type": "base64", "media_type": "application/pdf", "data": "…"}} {"type": "document", "source": {"type": "url", "url": "https://…"}}
/v1/responses {"type": "input_file", "file_data": "data:application/pdf;base64,…"} {"type": "input_file", "file_url": "https://…"}
{
  "model": "shannon-3",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "document",
          "source": {
            "type": "base64",
            "media_type": "application/pdf",
            "data": "JVBERi0xLjcKJeLjz9MK…"
          }
        },
        {"type": "text", "text": "Summarise this report in five points."}
      ]
    }
  ]
}
  • Nye media type ezigbo nke file ahụ. Na-enweghị ya, a na-agụ akwụkwọ dị ka application/pdf.
  • A na-agụ plain-text file, dị ka CSV, dị ka text ma jikọta ya n'otu ụzọ ahụ.
  • Mgbe enweghị ike ịgụ akwụkwọ, ma ọ bụ file ahụ bụ ụdị ọzọ, a na-agwa model ya ma ọ nwere ike ikwu ya n'azịza ya. Status ka bụ 200.
  • Na model ọ bụla ọzọ, a na-ahapụ document part n'ihe model na-agụ, a na-azakwa request site na ndị ọzọ.
  • A na-ebubata akwụkwọ e zipụrụ site na adreesị n'okpuru limits maka remote files dị n'elu.

Njehie

Status Ụdị Ozi Mgbe
400 invalid_request_error <id> does not accept image input Zipụrụ image na hosted open-weight model na-anara text naanị. A naghị ewepụ ihe ọ bụla na balance gị.
400 invalid_request_error audio and video input are not supported by any hosted model Zipụrụ audio ma ọ bụ video part na hosted open-weight model.
413 invalid_request_error Failed to buffer the request body: … Request body, ya na inline files ya, buru ibu karịa 32 MiB. Njehie ahụ nwere code: "request_too_large".

Shannon model na-aza na status 200 ihe ọ bụla files request nwere: a na-ahapụ part ọ naghị agụ. Njikwa njehie