Ma model ndi mitengo

ma model opanda zoletsa

22

context window

256K

kuyambira

$0.50 /1M

Ma hosted open-weight model

context ya ma token 262,144 pa id iliyonse; input ndi output zimawerengedwa mosiyana; id iliyonse imalandira zomwe model yomwe imatsanzira imalandira

Model	Input	Structured output	In / cached / out pa 1M	Zambiri
DeepSeek-V4-Pro-08133BIT-REAPDeepSeek · 1.6T MoE · 49B activeYopanda zoletsa	Text yolowa	JSON schema	$1.95 / $0.488 / $3.90	Zambiri
GLM-5.23BIT-REAPZ.ai · 744B MoE · 40B activeYopanda zoletsa	Text yolowa	JSON schema	$0.73 / $0.183 / $2.34	Zambiri
Kimi-K33BIT-REAPMoonshot AI · 2.8T MoE · 104B activeYopanda zoletsa	Text + chithunzi cholowa	JSON schema	$3.83 / $0.958 / $19.12	Zambiri
Nemotron3Ultra3BIT-REAPNVIDIA · 550B hybrid Mamba-Attention MoE · 55B activeYopanda zoletsa	Text yolowa	JSON schema	$0.75 / $0.188 / $3.30	Zambiri
MiniMax-M33BIT-REAPMiniMax · 428B MoE · 23B active · sparse attentionYopanda zoletsa	Text + chithunzi cholowa	JSON schema	$0.50 / $0.125 / $2.00	Zambiri
DeepSeek-V4-Flash-0731W4A16-AUTOROUND-REAPDeepSeek · 284B MoE · 13B activeYopanda zoletsa	Text yolowa	JSON schema	$0.50 / $0.125 / $2.00	Zambiri
Kimi-K2.6W4A16-AUTOROUND-REAPMoonshot AI · 1T MoE · 32B activeYopanda zoletsa	Text + chithunzi cholowa	JSON schema	$0.78 / $0.195 / $3.67	Zambiri
Laguna-S-2.1W4A16-AUTOROUND-REAPPoolside · 118B MoE · 8B activeYopanda zoletsa	Text yolowa	Palibe structured output	$0.50 / $0.125 / $2.00	Zambiri
inklingW4A16-AUTOROUND-REAPThinking Machines · 975B MoE · 41B activeYopanda zoletsa	Text + chithunzi cholowa	JSON object	$1.42 / $0.355 / $6.07	Zambiri
MiMo-V2.5-ProW8A16Xiaomi · 1.02T MoE · 42B active · 8-bitYopanda zoletsa	Text yolowa	JSON schema	$0.50 / $0.125 / $2.00	Zambiri
MiMo-V2.5W8A16Xiaomi · 8-bit multimodal · W8A16Yopanda zoletsa	Text + chithunzi cholowa	JSON schema	$0.50 / $0.125 / $2.00	Zambiri
Hy3W8A16Tencent · 295B MoE · 21B active · 8-bitYopanda zoletsa	Text yolowa	JSON schema	$0.50 / $0.125 / $2.00	Zambiri

Ma model a Shannon

mtengo umodzi wokhazikika wa input ndi output

Model	Context	Input	Structured output	Mtengo / ma token 1M
Shannon 1.6 Liteshannon-1.6-liteYopanda zoletsa	192K	Text + chithunzi cholowa	JSON schema	$3.90
Shannon 1.6 Proshannon-1.6-proYopanda zoletsa	192K	Text + chithunzi cholowa	JSON schema	$7.80
Shannon 2 Liteshannon-2-liteYopanda zoletsa	192K	Text + chithunzi cholowa	JSON schema	$3.90
Shannon 2 Proshannon-2-proYopanda zoletsa	192K	Text + chithunzi cholowa	JSON schema	$5.85
Shannon 3shannon-3Yopanda zoletsa	192K	Text + chithunzi cholowa	JSON schema	$3.35
Shannon 3 Proshannon-3-proYopanda zoletsa	192K	Text + chithunzi cholowa	JSON schema	$3.35
Shannon 3.1shannon-3.1Yopanda zoletsa	192K	Text + chithunzi cholowa	JSON schema	$3.35
Shannon 3.1 Proshannon-3.1-proYopanda zoletsa	192K	Text + chithunzi cholowa	JSON schema	$3.35
Shannon 3.5shannon-3.5Yopanda zoletsa	192K	Text + chithunzi cholowa	JSON schema	$3.35
Shannon Coder 1shannon-coder-1Yopanda zoletsa	128K	Text yolowa	JSON schema	$8.00

Pitani ku Quickstart Yesani Playground

Momwe mungawerengere mzere

Model iliyonse imachita stream ndipo imatha kuyitana ma tool. Zina za API zafotokozedwa patsamba la Capabilities. Mphamvu

Context window ndi malire a output

• Context window imaphatikiza chilichonse chomwe model imawerenga pa yankho limodzi: ma message anu, matanthauzo a ma tool ndi zithunzi.
• Banja la Shannon 3 limakwanitsa zokambirana zazitali paokha: limasunga ma turn atsopano kwambiri omwe akwana mu window ndipo limayankha.
• Mu stream, prompt yayitali kwambiri imanenedwa ngati error frame yokhala ndi type invalid_request_error ndi code: "context_length_exceeded".
• max_tokens imalandira mitengo kuyambira 1 mpaka 65,536 pa model iliyonse; max_output_tokens mu GET /v1/models ndi pamwamba pa malirewo. Default ndi 4,096.
• Pa model iliyonse, max_tokens ndi kuchuluka komwe pempho limasunga pa balance yanu.
• Ngati malire a kutalika kwa yankho, max_tokens imagwiritsidwa ntchito ndi hosted open-weight model, shannon-1.6-lite, shannon-1.6-pro ndi shannon-coder-1.
• Pa hosted open-weight model imalepheretsa text ya yankho. Reasoning yomwe ili yankho lisanafike siiwerengedwa mmenemo, ndipo mtengo wochepera 256 umagwira ntchito ngati 256.
• Stream ya shannon-1.6-lite, shannon-1.6-pro, shannon-coder-1 kapena banja la Shannon 3 imatha ndi finish_reason: "length" pamene yankho ladulidwa pa malire ake.

Malire ndi balance

Momwe mtengo umapangidwira

• Mitengo ndi mu USD pa ma token 1,000,000. Balance yanu imawerengedwa mu ma token ofunika $5.00 pa 1,000,000, kotero mtengo wa model umanena kuchuluka kwa balance komwe ma token ake amatenga.
• Ma model a Shannon ali ndi mtengo umodzi. Call imaperekedwa mtengo wa usage.total_tokens pa mtengo umenewo.
• Hosted open-weight model zili ndi mitengo itatu. Input yowerengedwa kuchokera ku prompt cache imaperekedwa mtengo wa cached, womwe ndi 25% ya mtengo wa input wozunguliridwa kufika $0.001. Zotsalira za input zimaperekedwa mtengo wa input, output pa mtengo wa output.
• Hosted open-weight model zimawerenga ma token ndi tokenizer ya model iliyonse. GET /v1/models imalemba izi ngati tokenizer: "original".
• shannon-coder-1 imawerengedwa mu ma call, osati ma token, pa /v1/chat/completions ndi /v1/messages. Mtengo wake wolembedwa umagwira ntchito pa /v1/responses.

Zomwe pempho limasunga, zitsanzo zowerengedwa za mtengo ndi allowance ya ma call a Shannon Coder zili pa Malire ndi balance. Malire ndi balance

Ma model id

Tumizani imodzi mwa ma id awa ngati 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

• Zilembo zazikulu ndi zazing'ono sizimakhudza, ndipo ma space ozungulira id amachotsedwa: SHANNON-3 ndi shannon-3 ndi model yomweyo.
• Yankho nthawi zonse limanyamula id yofalitsidwa mu model, mosasamala kalembedwe komwe mwatumiza.
• Dzina lowonetsera monga Shannon 3 si id.

Zolemba zinayi zazifupi zimalandiridwanso.

Id yomwe sinafalitsidwe imayankhidwa ndi status 400, type invalid_request_error ndi message unknown model: <id> pa ma endpoint onse atatu. Palibe model ina yomwe imayankha m'malo mwake.

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

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

• Nthawi zonse tumizani model. /v1/responses imayankha pempho lopanda iyo ndi 400. Pa /v1/chat/completions ndi /v1/messages pempho loterolo limayankhidwa ndi shannon-1.6-lite.
• POST /v1/tokenize ndi POST /v1/messages/count_tokens zimalandira ma id a hosted open-weight.

Kulemba ma model

GET /v1/models imabweza ma model a tsamba ili ngati JSON, ndi context window, mitengo ndi capabilities zomwezo. Igwiritseni ntchito kudzaza model picker kapena kuyang'ana mtengo mu 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"

Yankho, lofupikitsidwa ku entry imodzi ya mtundu uliwonse:

{
  "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 siifuna kiyi. Yankho ndi lofanana kwa aliyense woitana ndipo silidalira plan yanu.
• Mndandanda uli ndi ma id omwe angaitanidwe. Model yomwe siingatumikire pakali pano imasiyidwa.
• Palibe endpoint ya model imodzi: GET /v1/models/{id} imayankha 404 ndi type not_found_error.