Models & farashi

models marasa tacewa

22

context window

256K

daga

$0.50 /1M

Hosted open-weight models

context na tokens 262,144 a kowane id; ana lissafin input da output daban; kowane id yana karɓar abin da model ɗin da yake kwaikwaya yake karɓa

Model	Input	Structured output	In / cached / out kowace 1M	Ƙari
DeepSeek-V4-Pro-08133BIT-REAPDeepSeek · 1.6T MoE · 49B activeMara tacewa	Rubutu a input	JSON schema	$1.95 / $0.488 / $3.90	Ƙari
GLM-5.23BIT-REAPZ.ai · 744B MoE · 40B activeMara tacewa	Rubutu a input	JSON schema	$0.73 / $0.183 / $2.34	Ƙari
Kimi-K33BIT-REAPMoonshot AI · 2.8T MoE · 104B activeMara tacewa	Rubutu + hoto a input	JSON schema	$3.83 / $0.958 / $19.12	Ƙari
Nemotron3Ultra3BIT-REAPNVIDIA · 550B hybrid Mamba-Attention MoE · 55B activeMara tacewa	Rubutu a input	JSON schema	$0.75 / $0.188 / $3.30	Ƙari
MiniMax-M33BIT-REAPMiniMax · 428B MoE · 23B active · sparse attentionMara tacewa	Rubutu + hoto a input	JSON schema	$0.50 / $0.125 / $2.00	Ƙari
DeepSeek-V4-Flash-0731W4A16-AUTOROUND-REAPDeepSeek · 284B MoE · 13B activeMara tacewa	Rubutu a input	JSON schema	$0.50 / $0.125 / $2.00	Ƙari
Kimi-K2.6W4A16-AUTOROUND-REAPMoonshot AI · 1T MoE · 32B activeMara tacewa	Rubutu + hoto a input	JSON schema	$0.78 / $0.195 / $3.67	Ƙari
Laguna-S-2.1W4A16-AUTOROUND-REAPPoolside · 118B MoE · 8B activeMara tacewa	Rubutu a input	Babu structured output	$0.50 / $0.125 / $2.00	Ƙari
inklingW4A16-AUTOROUND-REAPThinking Machines · 975B MoE · 41B activeMara tacewa	Rubutu + hoto a input	JSON object	$1.42 / $0.355 / $6.07	Ƙari
MiMo-V2.5-ProW8A16Xiaomi · 1.02T MoE · 42B active · 8-bitMara tacewa	Rubutu a input	JSON schema	$0.50 / $0.125 / $2.00	Ƙari
MiMo-V2.5W8A16Xiaomi · 8-bit multimodal · W8A16Mara tacewa	Rubutu + hoto a input	JSON schema	$0.50 / $0.125 / $2.00	Ƙari
Hy3W8A16Tencent · 295B MoE · 21B active · 8-bitMara tacewa	Rubutu a input	JSON schema	$0.50 / $0.125 / $2.00	Ƙari

Models na Shannon

farashi tsayayye ɗaya don input da output

Model	Context	Input	Structured output	Farashi / 1M tokens
Shannon 1.6 Liteshannon-1.6-liteMara tacewa	192K	Rubutu + hoto a input	JSON schema	$3.90
Shannon 1.6 Proshannon-1.6-proMara tacewa	192K	Rubutu + hoto a input	JSON schema	$7.80
Shannon 2 Liteshannon-2-liteMara tacewa	192K	Rubutu + hoto a input	JSON schema	$3.90
Shannon 2 Proshannon-2-proMara tacewa	192K	Rubutu + hoto a input	JSON schema	$5.85
Shannon 3shannon-3Mara tacewa	192K	Rubutu + hoto a input	JSON schema	$3.35
Shannon 3 Proshannon-3-proMara tacewa	192K	Rubutu + hoto a input	JSON schema	$3.35
Shannon 3.1shannon-3.1Mara tacewa	192K	Rubutu + hoto a input	JSON schema	$3.35
Shannon 3.1 Proshannon-3.1-proMara tacewa	192K	Rubutu + hoto a input	JSON schema	$3.35
Shannon 3.5shannon-3.5Mara tacewa	192K	Rubutu + hoto a input	JSON schema	$3.35
Shannon Coder 1shannon-coder-1Mara tacewa	128K	Rubutu a input	JSON schema	$8.00

Tsallake zuwa Quickstart Gwada filin wasa

Yadda ake karanta layi

Kowane model yana yin streaming kuma yana iya kiran tools. Sauran fasalolin API an bayyana su a shafin Capabilities. Abubuwan iyawa

Context window da iyakokin output

• Context window yana ƙunshe da duk abin da model ke karantawa don amsa ɗaya: saƙonninku, ma'anar tools da hotuna.
• Iyalin Shannon 3 yana daidaita tattaunawa mafi tsawo da kansa: yana riƙe sabbin turns da suka shiga cikin taga kuma yana amsawa.
• A stream, prompt da ya yi tsawo da yawa ana bayyana shi a matsayin error frame mai nau'in invalid_request_error da code: "context_length_exceeded".
• max_tokens yana karɓar ƙimomi daga 1 zuwa 65,536 a kowane model; max_output_tokens a GET /v1/models shi ne saman wannan kewayon. Na asali shi ne 4,096.
• A kowane model, max_tokens shi ne adadin da request ke ware daga balance ɗinku.
• A matsayin iyaka kan tsawon amsa, hosted open-weight models, shannon-1.6-lite, shannon-1.6-pro da shannon-coder-1 ne ke aiwatar da max_tokens.
• A hosted open-weight models yana iyakance rubutun amsa. Ba a ƙirga reasoning kafin amsar a ciki ba, kuma ƙima ƙasa da 256 tana aiki kamar 256.
• Stream na shannon-1.6-lite, shannon-1.6-pro, shannon-coder-1 ko iyalin Shannon 3 yana ƙarewa da finish_reason: "length" idan an yanke amsar a iyakarta.

Iyakoki da balance

Yadda ake ƙirƙirar farashi

• Farashi suna a USD kowace tokens 1,000,000. Ana ƙirga balance ɗinku a tokens masu darajar $5.00 kowace 1,000,000, don haka farashin model yana faɗin yawan balance da tokens ɗinsa ke ɗauka.
• Models na Shannon suna da farashi ɗaya. Ana cajin kira usage.total_tokens a wannan farashi.
• Hosted open-weight models suna da farashi uku. Input da aka karanta daga prompt cache ana cajin sa a cached price, wato 25% na farashin input da aka zagaye zuwa $0.001. Ana cajin sauran input a farashin input, output a farashin output.
• Hosted open-weight models suna ƙirga tokens da tokenizer na kowane model. GET /v1/models yana alamta wannan a matsayin tokenizer: "original".
• Ana ƙirga shannon-coder-1 a kira, ba tokens ba, a /v1/chat/completions da /v1/messages. Farashinsa da aka lissafa yana aiki a /v1/responses.

Abin da request ke ware, misalan caji da allowance na kiran Shannon Coder suna a Iyakoki da balance. Iyakoki da balance

Model ids

Ku aika ɗaya daga cikin waɗannan ids a matsayin 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

• Manyan haruffa da ƙanana ba su da muhimmanci, kuma ana cire sarari a kewayen id: SHANNON-3 da shannon-3 model ɗaya ne.
• Amsa koyaushe tana ɗauke da id da aka wallafa a model, ko wane rubutu kuka aika.
• Sunan nuni kamar Shannon 3 ba id ba ne.

Ana kuma karɓar gajerun rubutu huɗu.

Id da ba a wallafa ba ana amsa shi da matsayi 400, nau'in invalid_request_error da saƙon unknown model: <id> a dukkan endpoints uku. Babu wani model da ke amsawa a madadinsa.

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

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

• Koyaushe ku aika model. /v1/responses yana amsa request ba tare da shi ba da 400. A /v1/chat/completions da /v1/messages shannon-1.6-lite ne ke amsa irin wannan request.
• POST /v1/tokenize da POST /v1/messages/count_tokens suna karɓar ids na hosted open-weight.

Jerin models

GET /v1/models yana mayar da models na wannan shafin a matsayin JSON, tare da context windows, farashi da capabilities iri ɗaya. Ku yi amfani da shi don cika model picker ko don duba farashi a cikin 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"

Amsar, an taƙaita zuwa shigarwa ɗaya na kowane iri:

{
  "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 ba ya buƙatar maɓalli. Amsar iri ɗaya ce ga kowane mai kira kuma ba ta dogara da plan ɗinku ba.
• Jerin yana ɗauke da ids da za a iya kira. Model da ba za a iya yi masa hidima a halin yanzu ba ana bar shi daga jerin.
• Babu endpoint na model ɗaya: GET /v1/models/{id} yana amsawa da 404 da nau'in not_found_error.