Immaġini u fajls
Ibgħat immaġini ma' messaġġ: liema mudelli jaċċettawhom, il-forom aċċettati u l-limiti tagħhom.
POST https://api.shannon-ai.com/v1/chat/completions
Immaġni tivvjaġġa ġewwa messaġġ tal-utent, bħala parti waħda mill-content tiegħu. L-eżempju jaqra fajl lokali u jibgħatu bħala 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
}
} Liema mudelli jaċċettaw immaġini
| Mudelli | Jaqra |
|---|---|
shannon-3, shannon-3-pro, shannon-3.1, shannon-3.1-pro | Immaġini, u dokumenti: fajls PDF, Word, PowerPoint, Excel u tat-test sempliċi. |
shannon-1.6-lite, shannon-1.6-pro | Immaġini. |
shannon-2-lite, shannon-2-pro | Immaġini, f'talbiet li jibagħtu wkoll tools jew response_format. |
Kimi-K3-3BIT-REAP, MiniMax-M3-3BIT-REAP, Kimi-K2.6-W4A16-AUTOROUND-REAP, inkling-W4A16-AUTOROUND-REAP, MiMo-V2.5-W8A16 | Immaġini. |
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 | Test. Talba b'immaġni titwieġeb bi status 400. |
shannon-coder-1 | Test. |
GET /v1/models jirrapporta input ta' immaġini għal kull id bħala capabilities.vision. Mudelli u prezzijiet
- Il-mudelli Shannon jaqraw il-fajls tal-aħħar messaġġ tal-utent. Poġġi l-immaġni fil-messaġġ li jistaqsi dwarha.
- Il-familja Shannon 3 iżżomm f'għajnejha wkoll ftit immaġini u dokumenti ta' messaġġi preċedenti tal-utent meta ntbagħtu inline, u taqra massimu stabbilit ta' immaġini għal kull talba, magħdud mill-ewwel.
- Il-mudelli open-weight ospitati jaqraw l-immaġini ta' kull messaġġ tal-konversazzjoni.
Forom aċċettati
Fajl jintbagħat b'waħda minn żewġ modi: ġewwa t-talba bħala base64, jew bħala indirizz li l-API tiġbed. Kull fajl jivvjaġġa mat-talba; l-API m'għandhiex endpoint għat-tgħabbija.
| Endpoint | Fit-talba (base64) | Bl-indirizz |
|---|---|---|
/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://…"}} |
L-istess messaġġ tal-utent f'kull format:
{
"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 għandha l-forma
data:<media type>;base64,<data>. Il-parti;base64hija meħtieġa. - Fuq
/v1/chat/completionsu/v1/responses,image_urljista' jkun l-istring innifisha jew oġġett{"url": "…"}. - It-tip tal-fajl jiġi mid-data URL jew minn
media_type. Mingħajru, immaġni tinqara bħalaimage/png. - Il-fajls inline jingħaddu fid-daqs tal-body tat-talba, li jista' jkun sa 32 MiB.
- Aċċettat għall-kompatibbiltà:
image_url.detail.
Limiti għall-fajls remoti
Meta tibgħat indirizz, l-API tiġbed il-fajl qabel ma jaħdem il-mudell, f'dawn il-limiti:
| Limitu | Regola |
|---|---|
| Daqs | Sa 8 MiB għal kull fajl. |
| Ħin | 20 sekonda biex iwieġeb. |
| Redirects | Massimu ta' 5. |
| Indirizz | http:// jew https://, mingħajr isem tal-utent jew password fl-indirizz, fuq host b'indirizz pubbliku. |
| Tweġiba tal-host | Status fil-medda 200 u body li mhuwiex vojt. It-tip tal-fajl jinqara minn Content-Type. |
| Mitlub bħala | User-Agent: ShannonBot/1.0 (+https://shannon-ai.com) |
Kif jingħaddu l-immaġini f'tokens
Fuq il-mudelli open-weight ospitati immaġni tgħodd token wieħed għal kull patch ta' 28 × 28 pixel: il-wisa' diviża b'28, mqarrba 'l fuq, immultiplikata bl-għoli diviż b'28, mqarrab 'l fuq. Ir-riżultat huwa parti minn usage.prompt_tokens u jitħallas bħala input.
- 1024 × 1024
- 1,369 tokens
- 1920 × 1080
- 2,691 tokens
- 512 × 512
- 361 tokens
- Immaġni remota tinġieb l-ewwel, għalhekk jingħadd id-daqs veru tagħha.
- Immaġni li d-daqs tagħha ma jistax jinqara tgħodd 1,024 token.
- L-endpoints tal-għadd jużaw l-istess regola għad-data URLs. Ma jġibux indirizz remot u jgħoddu bħala 1,024 token. Għadd ta' tokens
- Għal mudell Shannon, aqra x'ħalset talba b'immaġini minn
usagefit-tweġiba.
Dokumenti u fajls oħra
Il-familja Shannon 3 (shannon-3, shannon-3-pro, shannon-3.1, shannon-3.1-pro) taqra wkoll dokumenti. It-test tad-dokument jinġibed u jingħata lill-mudell mal-messaġġ tiegħek.
| Dokument | Tip ta' media |
|---|---|
application/pdf | |
| Word | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| PowerPoint | application/vnd.openxmlformats-officedocument.presentationml.presentation |
| Excel | application/vnd.openxmlformats-officedocument.spreadsheetml.sheet |
Dokument huwa parti mill-messaġġ tal-utent, bħal immaġni:
| Endpoint | Fit-talba (base64) | Bl-indirizz |
|---|---|---|
/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."}
]
}
]
} - Agħti t-tip ta' media veru tal-fajl. Mingħajru, dokument jinqara bħala
application/pdf. - Fajl tat-test sempliċi, bħal CSV, jinqara bħala test u jiġi mehmuż bl-istess mod.
- Meta dokument ma jistax jinqara, jew il-fajl huwa ta' tip ieħor, il-mudell jingħata x'jifhem u jista' jgħidu fit-tweġiba tiegħu. L-istatus xorta huwa
200. - Fuq kull mudell ieħor parti ta' dokument titħalla barra minn dak li jaqra l-mudell, u t-talba titwieġeb mill-bqija.
- Dokument mibgħut bl-indirizz jinġieb bil-limiti għall-fajls remoti hawn fuq.
Żbalji
| Status | Tip | Messaġġ | Meta |
|---|---|---|---|
400 | invalid_request_error | <id> does not accept image input | Intbagħtet immaġni lil mudell open-weight ospitat li jaċċetta test biss. Ma jittieħed xejn mill-bilanċ tiegħek. |
400 | invalid_request_error | audio and video input are not supported by any hosted model | Parti awdjo jew vidjo ntbagħtet lil mudell open-weight ospitat. |
413 | invalid_request_error | Failed to buffer the request body: … | Il-body tat-talba, bil-fajls inline tiegħu, huwa akbar minn 32 MiB. L-iżball iġorr code: "request_too_large". |
Mudell Shannon iwieġeb bi status 200 ikun xi jkun il-fajls li ġġorr it-talba: parti li ma jaqrax titħalla barra. Ġestjoni tal-iżbalji