Sawirro iyo faylal
Sawirro la fariin dir: model-yada qaata, qaababka la aqbalo iyo xadadkooda.
POST https://api.shannon-ai.com/v1/chat/completions
Sawir wuxuu ku socdaa fariin isticmaale dhexdeeda, sida qayb ka mid ah content kiisa. Tusaalaha wuxuu akhriyaa fayl maxalli ah oo u diraa sida 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) import fs from "node:fs";
import OpenAI from "openai";
const client = new OpenAI({ apiKey: "YOUR_API_KEY", baseURL: "https://api.shannon-ai.com/v1" });
const image = fs.readFileSync("photo.jpg").toString("base64");
const response = await 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 } },
],
}],
});
console.log(response.choices[0].message.content); # Linux: base64 -w0 photo.jpg macOS: base64 -i photo.jpg
IMAGE=$(base64 -w0 photo.jpg)
curl https://api.shannon-ai.com/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d @- <<EOF
{
"model": "shannon-3",
"messages": [{
"role": "user",
"content": [
{"type": "text", "text": "What is in this picture?"},
{"type": "image_url", "image_url": {"url": "data:image/jpeg;base64,${IMAGE}"}}
]
}]
}
EOF {
"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
}
} Model-yada qaata sawirro
| Model-yo | Akhriyaa |
|---|---|
shannon-3, shannon-3-pro, shannon-3.1, shannon-3.1-pro | Sawirro, iyo dukumentiyo: PDF, Word, PowerPoint, Excel iyo faylal qoraal cad ah. |
shannon-1.6-lite, shannon-1.6-pro | Sawirro. |
shannon-2-lite, shannon-2-pro | Sawirro, codsiyada sidoo kale diraya tools ama response_format. |
Kimi-K3-3BIT-REAP, MiniMax-M3-3BIT-REAP, Kimi-K2.6-W4A16-AUTOROUND-REAP, inkling-W4A16-AUTOROUND-REAP, MiMo-V2.5-W8A16 | Sawirro. |
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 | Qoraal. Codsi leh sawir waxaa lagu jawaabaa xaalad 400. |
shannon-coder-1 | Qoraal. |
GET /v1/models wuxuu gelinta sawirka id kasta u sheegaa sida capabilities.vision. Model-yo & qiimo
- Model-yada Shannon waxay akhriyaan faylasha fariinta isticmaalaha ee ugu dambeysay. Sawirka geli fariinta ka weydiisaa.
- Qoyska Shannon 3 wuxuu sidoo kale ku hayaa muuqaal dhowr sawir iyo dukumenti oo fariimo isticmaale oo hore ah marka loo diray gudaha, wuxuuna akhriyaa tiro ugu badan oo go'an oo sawirro ah codsi kasta, laga tiriyo kan ugu horreeya.
- Model-yada open-weight ee la martigeliyo waxay akhriyaan sawirrada fariin kasta oo wada-hadalka ah.
Qaababka la aqbalo
Fayl waxaa loo diraa laba hab midkood: codsiga dhexdiisa sida base64, ama sida cinwaan API-gu soo qaado. Fayl kastaa wuxuu la socdaa codsiga; API-gu endpoint upload ma laha.
| Endpoint | Codsiga dhexdiisa (base64) | Cinwaan |
|---|---|---|
/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://…"}} |
Isla fariinta isticmaalaha qaab kasta:
{
"role": "user",
"content": [
{"type": "text", "text": "What is in this picture?"},
{
"type": "image_url",
"image_url": {
"url": "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD…"
}
}
]
} {
"role": "user",
"content": [
{"type": "input_text", "text": "What is in this picture?"},
{
"type": "input_image",
"image_url": "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD…"
}
]
} {
"role": "user",
"content": [
{"type": "text", "text": "What is in this picture?"},
{
"type": "image",
"source": {
"type": "base64",
"media_type": "image/jpeg",
"data": "/9j/4AAQSkZJRgABAQAAAQABAAD…"
}
}
]
} - Data URL waxay leedahay qaabka
data:<media type>;base64,<data>. Qaybta;base64waa loo baahan yahay. /v1/chat/completionsiyo/v1/responses,image_urlwuxuu noqon karaa string-ka laftiisa ama shay{"url": "…"}.- Nooca faylka wuxuu ka yimaadaa data URL ama
media_type. La'aantiis, sawir waxaa loo akhriyaa sidaimage/png. - Faylasha codsiga dhexdiisa ku jira waxay ka tirsan yihiin cabbirka jidhka codsiga, kaas oo noqon kara ilaa 32 MiB.
- La aqbalay u waafaqsanaan:
image_url.detail.
Xadadka faylasha fogaanta ah
Marka aad cinwaan dirto, API-gu faylka wuu soo qaadaa ka hor inta model-ku orodin, xadadkan gudahooda:
| Xad | Xeer |
|---|---|
| Cabbir | Ilaa 8 MiB fayl kasta. |
| Wakhti | 20 ilbiriqsi si uu u jawaabo. |
| Wareejin | Ugu badnaan 5. |
| Cinwaan | http:// ama https://, cinwaanka magac isticmaale ama password la'aantiis, host leh cinwaan dadweyne. |
| Jawaabta host-ka | Xaalad kala duwanaanshaha 200 ah iyo jidh aan madhnayn. Nooca faylka waxaa laga akhriyaa Content-Type. |
| La codsaday sida | User-Agent: ShannonBot/1.0 (+https://shannon-ai.com) |
Sida sawirrada loogu tiriyo token-yo
Model-yada open-weight ee la martigeliyo sawir wuxuu tiriyaa hal token patch kasta oo 28 × 28 pixel ah: ballaadhka loo qaybiyay 28, kor loo wareejiyay, oo lagu dhufto dhererka loo qaybiyay 28, kor loo wareejiyay. Natiijadu waxay qayb ka tahay usage.prompt_tokens waxaana lagu dallacaa sida input.
- 1024 × 1024
- 1,369 token-yo
- 1920 × 1080
- 2,691 token-yo
- 512 × 512
- 361 token-yo
- Sawir fogaan ah marka hore ayaa la soo qaadaa, sidaas darteed cabbirkiisa runta ah ayaa la tiriyaa.
- Sawir cabbirkiisa la akhrin karin wuxuu tirsan yahay 1,024 token.
- Endpoint-yada tirintu waxay isticmaalaan isla xeerkan data URL-yada. Cinwaan fogaan ah ma soo qaadaan oo waxay u tiriyaan 1,024 token. Tirinta token-yada
- Model Shannon, akhri waxa codsi leh sawirro ku kacay
usageee jawaabta.
Dukumentiyo iyo faylal kale
Qoyska Shannon 3 (shannon-3, shannon-3-pro, shannon-3.1, shannon-3.1-pro) wuxuu akhriyaa dukumentiyo sidoo kale. Qoraalka dukumentiga waa laga soo saaraa oo model-ka waa la siiyaa fariintaada.
| Dukumenti | Nooca warbaahinta |
|---|---|
application/pdf | |
| Word | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| PowerPoint | application/vnd.openxmlformats-officedocument.presentationml.presentation |
| Excel | application/vnd.openxmlformats-officedocument.spreadsheetml.sheet |
Dukumenti waa qayb ka mid ah fariinta isticmaalaha, sida sawir:
| Endpoint | Codsiga dhexdiisa (base64) | Cinwaan |
|---|---|---|
/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."}
]
}
]
} {
"model": "shannon-3",
"input": [
{
"role": "user",
"content": [
{
"type": "input_file",
"file_data": "data:application/pdf;base64,JVBERi0xLjcKJeLjz9MK…"
},
{
"type": "input_text",
"text": "Summarise this report in five points."
}
]
}
]
} {
"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."}
]
}
]
} - Sii nooca warbaahinta ee runta ah ee faylka. La'aantiis, dukumenti waxaa loo akhriyaa sida
application/pdf. - Fayl qoraal cad ah, sida CSV, waxaa loo akhriyaa sida qoraal oo si isku mid ah ayaa loo lifaaqaa.
- Marka dukumenti la akhrin karin, ama faylku yahay nooc kale, model-ka waa loo sheegaa wuxuuna ku sheegi karaa jawaabtiisa. Xaaladdu weli waa
200. - Model kasta oo kale qayb dukumenti ah waxaa laga tagaa waxa model-ku akhriyo, codsiga waxaana lagu jawaabaa inta kale.
- Dukumenti cinwaan lagu diray waa la soo qaadaa iyadoo la raacayo xadadka faylasha fogaanta ah ee kor ku xusan.
Khaladaad
| Xaalad | Nooc | Fariin | Goorta |
|---|---|---|---|
400 | invalid_request_error | <id> does not accept image input | Sawir ayaa loo diray model open-weight oo la martigeliyay oo qaata qoraal oo keliya. Haraaggaaga wax laga ma qaado. |
400 | invalid_request_error | audio and video input are not supported by any hosted model | Qayb cod ama muuqaal ah ayaa loo diray model open-weight oo la martigeliyay. |
413 | invalid_request_error | Failed to buffer the request body: … | Jidhka codsiga, iyo faylasha dhexdiisa ku jira, wuu ka weyn yahay 32 MiB. Khaladku wuxuu sidaa code: "request_too_large". |
Model Shannon wuxuu ku jawaabaa xaalad 200 fayl kasta oo codsigu sido: qayb uusan akhrin laga tagaa. Khaladaadka Maareynta