Model na bei

model zisizodhibitiwa

22

dirisha la muktadha

256K

kuanzia

$0.50 /1M

Model za open-weight zinazopangishwa

muktadha wa tokens 262,144 kwenye kila id; input na output zinatozwa kando; kila id inakubali kile kinachokubaliwa na model inayoiga

Model	Input	Matokeo yaliyopangwa	Ndani / cached / nje kwa 1M	Zaidi
DeepSeek-V4-Pro-08133BIT-REAPDeepSeek · 1.6T MoE · 49B activeIsiyodhibitiwa	Maandishi kama input	JSON schema	$1.95 / $0.488 / $3.90	Zaidi
GLM-5.23BIT-REAPZ.ai · 744B MoE · 40B activeIsiyodhibitiwa	Maandishi kama input	JSON schema	$0.73 / $0.183 / $2.34	Zaidi
Kimi-K33BIT-REAPMoonshot AI · 2.8T MoE · 104B activeIsiyodhibitiwa	Maandishi + picha kama input	JSON schema	$3.83 / $0.958 / $19.12	Zaidi
Nemotron3Ultra3BIT-REAPNVIDIA · 550B hybrid Mamba-Attention MoE · 55B activeIsiyodhibitiwa	Maandishi kama input	JSON schema	$0.75 / $0.188 / $3.30	Zaidi
MiniMax-M33BIT-REAPMiniMax · 428B MoE · 23B active · sparse attentionIsiyodhibitiwa	Maandishi + picha kama input	JSON schema	$0.50 / $0.125 / $2.00	Zaidi
DeepSeek-V4-Flash-0731W4A16-AUTOROUND-REAPDeepSeek · 284B MoE · 13B activeIsiyodhibitiwa	Maandishi kama input	JSON schema	$0.50 / $0.125 / $2.00	Zaidi
Kimi-K2.6W4A16-AUTOROUND-REAPMoonshot AI · 1T MoE · 32B activeIsiyodhibitiwa	Maandishi + picha kama input	JSON schema	$0.78 / $0.195 / $3.67	Zaidi
Laguna-S-2.1W4A16-AUTOROUND-REAPPoolside · 118B MoE · 8B activeIsiyodhibitiwa	Maandishi kama input	Hakuna matokeo yaliyopangwa	$0.50 / $0.125 / $2.00	Zaidi
inklingW4A16-AUTOROUND-REAPThinking Machines · 975B MoE · 41B activeIsiyodhibitiwa	Maandishi + picha kama input	Kitu cha JSON	$1.42 / $0.355 / $6.07	Zaidi
MiMo-V2.5-ProW8A16Xiaomi · 1.02T MoE · 42B active · 8-bitIsiyodhibitiwa	Maandishi kama input	JSON schema	$0.50 / $0.125 / $2.00	Zaidi
MiMo-V2.5W8A16Xiaomi · 8-bit multimodal · W8A16Isiyodhibitiwa	Maandishi + picha kama input	JSON schema	$0.50 / $0.125 / $2.00	Zaidi
Hy3W8A16Tencent · 295B MoE · 21B active · 8-bitIsiyodhibitiwa	Maandishi kama input	JSON schema	$0.50 / $0.125 / $2.00	Zaidi

Model za Shannon

bei moja isiyobadilika kwa input na output

Model	Muktadha	Input	Matokeo yaliyopangwa	Bei / tokens 1M
Shannon 1.6 Liteshannon-1.6-liteIsiyodhibitiwa	192K	Maandishi + picha kama input	JSON schema	$3.90
Shannon 1.6 Proshannon-1.6-proIsiyodhibitiwa	192K	Maandishi + picha kama input	JSON schema	$7.80
Shannon 2 Liteshannon-2-liteIsiyodhibitiwa	192K	Maandishi + picha kama input	JSON schema	$3.90
Shannon 2 Proshannon-2-proIsiyodhibitiwa	192K	Maandishi + picha kama input	JSON schema	$5.85
Shannon 3shannon-3Isiyodhibitiwa	192K	Maandishi + picha kama input	JSON schema	$3.35
Shannon 3 Proshannon-3-proIsiyodhibitiwa	192K	Maandishi + picha kama input	JSON schema	$3.35
Shannon 3.1shannon-3.1Isiyodhibitiwa	192K	Maandishi + picha kama input	JSON schema	$3.35
Shannon 3.1 Proshannon-3.1-proIsiyodhibitiwa	192K	Maandishi + picha kama input	JSON schema	$3.35
Shannon 3.5shannon-3.5Isiyodhibitiwa	192K	Maandishi + picha kama input	JSON schema	$3.35
Shannon Coder 1shannon-coder-1Isiyodhibitiwa	128K	Maandishi kama input	JSON schema	$8.00

Nenda kwenye Mwanzo wa Haraka Jaribu Playground

Jinsi ya kusoma safu

Kila model hufanya stream na inaweza kuita tools. Vipengele vingine vya API vimeelezwa kwenye ukurasa wa Uwezo. Uwezo

Dirisha la muktadha na mipaka ya output

• Dirisha la muktadha linafunika kila kitu ambacho model husoma kwa jibu moja: ujumbe wako, ufafanuzi wa tools na picha.
• Familia ya Shannon 3 hutoshea mazungumzo marefu zaidi yenyewe: huweka zamu mpya zaidi zinazotoshea kwenye dirisha na kujibu.
• Kwenye stream, prompt ndefu mno huripotiwa kama frame ya kosa yenye aina invalid_request_error na code: "context_length_exceeded".
• max_tokens inakubali thamani kutoka 1 hadi 65,536 kwenye kila model; max_output_tokens kwenye GET /v1/models ni kilele cha masafa hayo. Chaguo-msingi ni 4,096.
• Kwenye kila model, max_tokens ni kiasi ambacho ombi hutenga kutoka kwenye salio lako.
• Kama kikomo cha urefu wa jibu, max_tokens inatumiwa na model za open-weight zinazopangishwa, shannon-1.6-lite, shannon-1.6-pro na shannon-coder-1.
• Kwenye model za open-weight zinazopangishwa inapunguza maandishi ya jibu. Reasoning kabla ya jibu haihesabiwi dhidi yake, na thamani chini ya 256 hufanya kazi kama 256.
• Stream ya shannon-1.6-lite, shannon-1.6-pro, shannon-coder-1 au familia ya Shannon 3 huisha na finish_reason: "length" jibu likikatwa kwenye kikomo chake.

Mipaka na salio

Jinsi bei inavyoundwa

• Bei ziko kwa USD kwa tokens 1,000,000. Salio lako linahesabiwa kwa tokens zenye thamani ya $5.00 kwa 1,000,000, kwa hiyo bei ya model inasema tokens zake zinachukua salio kiasi gani.
• Model za Shannon zina bei moja. Wito hutozwa usage.total_tokens kwa bei hiyo.
• Model za open-weight zinazopangishwa zina bei tatu. Input inayosomwa kutoka prompt cache hutozwa kwa bei ya cached, ambayo ni 25% ya bei ya input iliyozungushwa hadi $0.001. Input iliyobaki hutozwa kwa bei ya input, output kwa bei ya output.
• Model za open-weight zinazopangishwa huhesabu tokens kwa tokenizer ya kila model. GET /v1/models inaonyesha hili kama tokenizer: "original".
• shannon-coder-1 huhesabiwa kwa wito, si tokens, kwenye /v1/chat/completions na /v1/messages. Bei yake iliyoorodheshwa inatumika kwenye /v1/responses.

Ombi hutenga nini, mifano iliyokokotolewa ya gharama na kiasi cha wito wa Shannon Coder viko kwenye Mipaka na salio. Mipaka na salio

Model id

Tuma mojawapo ya id hizi kama model.

shannon-1.6-lite
shannon-1.6-pro
shannon-2-lite
shannon-2-pro
shannon-3
shannon-3-pro
shannon-3.1
shannon-3.1-pro
shannon-3.5
shannon-coder-1
DeepSeek-V4-Pro-0813-3BIT-REAP
GLM-5.2-3BIT-REAP
Kimi-K3-3BIT-REAP
Nemotron3Ultra-3BIT-REAP
MiniMax-M3-3BIT-REAP
DeepSeek-V4-Flash-0731-W4A16-AUTOROUND-REAP
Kimi-K2.6-W4A16-AUTOROUND-REAP
Laguna-S-2.1-W4A16-AUTOROUND-REAP
inkling-W4A16-AUTOROUND-REAP
MiMo-V2.5-Pro-W8A16
MiMo-V2.5-W8A16
Hy3-W8A16

• Herufi kubwa na ndogo hazijalishi, na nafasi kuzunguka id huondolewa: SHANNON-3 na shannon-3 ni model ile ile.
• Jibu daima hubeba id iliyochapishwa kwenye model, bila kujali tahajia uliyotuma.
• Jina la kuonyesha kama Shannon 3 si id.

Tahajia nne fupi zaidi pia zinakubaliwa.

Id ambayo haijachapishwa hujibiwa kwa hali 400, aina invalid_request_error na ujumbe unknown model: <id> kwenye endpoint zote tatu. Hakuna model nyingine inayojibu badala yake.

{
  "error": {
    "type": "invalid_request_error",
    "message": "unknown model: gpt-4o"
  }
}

{
  "type": "error",
  "error": {
    "type": "invalid_request_error",
    "message": "unknown model: gpt-4o"
  }
}

• Tuma model daima. /v1/responses hujibu ombi lisilo nayo kwa 400. Kwenye /v1/chat/completions na /v1/messages ombi kama hilo hujibiwa na shannon-1.6-lite.
• POST /v1/tokenize na POST /v1/messages/count_tokens zinakubali id za open-weight zinazopangishwa.

Orodhesha model

GET /v1/models hurudisha model za ukurasa huu kama JSON, zikiwa na madirisha ya muktadha, bei na capabilities zile zile. Itumie kujaza model picker au kukagua bei kwenye code.

GET https://api.shannon-ai.com/v1/models

from openai import OpenAI

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

for model in client.models.list():
    print(model.id)

import OpenAI from 'openai';

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

for await (const model of client.models.list()) {
  console.log(model.id);
}

curl "https://api.shannon-ai.com/v1/models"

Jibu, limefupishwa hadi entry moja ya kila aina:

{
  "object": "list",
  "data": [
    {
      "id": "shannon-3",
      "object": "model",
      "created": 1700000000,
      "owned_by": "shannon-ai",
      "display_name": "Shannon 3",
      "context_window": 196608,
      "max_output_tokens": 65536,
      "pricing": {
        "input_per_million_usd": 3.35,
        "output_per_million_usd": 3.35,
        "currency": "USD"
      },
      "capabilities": {
        "tools": true,
        "vision": true,
        "reasoning": true,
        "reasoning_effort": false,
        "streaming": true,
        "json_schema": true
      },
      "endpoints": [
        "/v1/chat/completions",
        "/v1/messages",
        "/v1/responses"
      ]
    },
    {
      "id": "MiniMax-M3-3BIT-REAP",
      "object": "model",
      "created": 1700000000,
      "owned_by": "shannon-ai",
      "display_name": "MiniMax-M3 · 3BIT-REAP",
      "context_window": 262144,
      "max_output_tokens": 65536,
      "pricing": {
        "input_per_million_usd": 0.5,
        "cached_input_per_million_usd": 0.125,
        "output_per_million_usd": 2,
        "currency": "USD"
      },
      "tokenizer": "original",
      "capabilities": {
        "tools": true,
        "vision": true,
        "reasoning": true,
        "reasoning_effort": true,
        "streaming": true,
        "response_format": true,
        "json_schema": true,
        "prompt_caching": true
      },
      "endpoints": [
        "/v1/chat/completions",
        "/v1/messages",
        "/v1/responses",
        "/v1/messages/count_tokens",
        "/v1/tokenize"
      ]
    }
  ]
}

• GET /v1/models haihitaji key. Jibu ni lile lile kwa kila mwitaji na halitegemei mpango wako.
• Orodha ina id zinazoweza kuitwa. Model isiyoweza kuhudumiwa kwa sasa huachwa nje yake.
• Hakuna endpoint ya model moja: GET /v1/models/{id} hujibu 404 na aina not_found_error.