Dukumentiga API
Kaabayaasha Sirdoon Aan La Faragelin
API-ga AI-ga aan la faragelin: model-yada Shannon oo aan diidmo iyo shaandhayn lahayn, laga bixiyo GPU-yadayada. Waafaqsan OpenAI iyo Anthropic — geli SDK-ga hore u isticmaalayso, hal fure, streaming, tool calling iyo reasoning endpoint kasta.
- 22 model aan la faragelin
- 256K daaqadda context-ka
- laga bilaabo $0.50/1M
- 3 API is-waafaqsan
- 3 shabakado: TLS, Tor, I2P
Bilow dhismaha Isku day garoonka ciyaaraha
from openai import OpenAI
client = OpenAI(
api_key="YOUR_API_KEY",
base_url="https://api.shannon-ai.com/v1"
)
response = client.chat.completions.create(
model="shannon-3",
messages=[
{"role": "user", "content": "Hello, Shannon!"}
],
max_tokens=1024
)
print(response.choices[0].message.content) import OpenAI from 'openai';
const client = new OpenAI({
apiKey: 'YOUR_API_KEY',
baseURL: 'https://api.shannon-ai.com/v1'
});
const response = await client.chat.completions.create({
model: 'shannon-3',
messages: [
{ role: 'user', content: 'Hello, Shannon!' }
],
max_tokens: 1024
});
console.log(response.choices[0].message.content); package main
import (
"context"
"fmt"
openai "github.com/sashabaranov/go-openai"
)
func main() {
config := openai.DefaultConfig("YOUR_API_KEY")
config.BaseURL = "https://api.shannon-ai.com/v1"
client := openai.NewClientWithConfig(config)
resp, err := client.CreateChatCompletion(
context.Background(),
openai.ChatCompletionRequest{
Model: "shannon-3",
Messages: []openai.ChatCompletionMessage{
{Role: "user", Content: "Hello, Shannon!"},
},
MaxTokens: 1024,
},
)
if err != nil {
panic(err)
}
fmt.Println(resp.Choices[0].Message.Content)
} curl -X POST "https://api.shannon-ai.com/v1/chat/completions" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "shannon-3",
"messages": [
{"role": "user", "content": "Hello, Shannon!"}
],
"max_tokens": 1024
}' Model-yo & qiimo HALKII 1M TOKEN
Xadka, id-yada iyo sida qiimaha loo sameeyoModel-yada open-weight ee la martigeliyo 262,144-token context id kasta; input iyo output si gooni ah ayaa loo xisaabiyaa; id kasta wuxuu aqbalaa wixii model-ka uu ku dayanayo aqbalo
- Aan la faragelin
DeepSeek-V4-Pro-0813
3BIT-REAPDeepSeek · 1.6T MoE · 49B active
Model cabbir flagship ah oo loogu talagalay reasoning qoto dheer iyo wax-soo-saar dheer.
Id-ga model-kaDeepSeek-V4-Pro-0813-3BIT-REAP- Context
- 256K
- Input
- Qoraal gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $1.95 / $0.488 / $3.90
- Aan la faragelin
GLM-5.2
3BIT-REAPZ.ai · 744B MoE · 40B active
Model guud oo xoog leh oo leh wax-soo-saar luqado badan oo hodan ah.
Id-ga model-kaGLM-5.2-3BIT-REAP- Context
- 256K
- Input
- Qoraal gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $0.73 / $0.183 / $2.34
- Aan la faragelin
Kimi-K3
3BIT-REAPMoonshot AI · 2.8T MoE · 104B active
Model heer sare ah oo loo habeeyay jawaabo dhamaystiran oo faahfaahsan.
Id-ga model-kaKimi-K3-3BIT-REAP- Context
- 256K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $3.83 / $0.958 / $19.12
- Aan la faragelin
Nemotron3Ultra
3BIT-REAPNVIDIA · 550B hybrid Mamba-Attention MoE · 55B active
Model weyn oo hufan oo leh reasoning xoog leh halkii doolar.
Id-ga model-kaNemotron3Ultra-3BIT-REAP- Context
- 256K
- Input
- Qoraal gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $0.75 / $0.188 / $3.30
- Aan la faragelin
MiniMax-M3
3BIT-REAPMiniMax · 428B MoE · 23B active · sparse attention
Model degdeg ah oo jaban oo loogu talagalay culeys badan.
Id-ga model-kaMiniMax-M3-3BIT-REAP- Context
- 256K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $0.50 / $0.125 / $2.00
- Aan la faragelin
DeepSeek-V4-Flash-0731
W4A16-AUTOROUND-REAPDeepSeek · 284B MoE · 13B active
Model DeepSeek oo degdeg ah oo isu dheelitira xawaaraha iyo tayada.
Id-ga model-kaDeepSeek-V4-Flash-0731-W4A16-AUTOROUND-REAP- Context
- 256K
- Input
- Qoraal gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $0.50 / $0.125 / $2.00
- Aan la faragelin
Kimi-K2.6
W4A16-AUTOROUND-REAPMoonshot AI · 1T MoE · 32B active
Model awood leh oo loogu talagalay reasoning iyo qorid maalinle ah.
Id-ga model-kaKimi-K2.6-W4A16-AUTOROUND-REAP- Context
- 256K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $0.78 / $0.195 / $3.67
- Aan la faragelin
Laguna-S-2.1
W4A16-AUTOROUND-REAPPoolside · 118B MoE · 8B active
Model aad u jaban oo loogu talagalay hawlo fudud.
Id-ga model-kaLaguna-S-2.1-W4A16-AUTOROUND-REAP- Context
- 256K
- Input
- Qoraal gelinaya
- Structured output
- Structured output ma jiro
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $0.50 / $0.125 / $2.00
- Aan la faragelin
inkling
W4A16-AUTOROUND-REAPThinking Machines · 975B MoE · 41B active
Model hal-abuur ah oo leh codka qorista u gaar ah.
Id-ga model-kainkling-W4A16-AUTOROUND-REAP- Context
- 256K
- Input
- Qoraal + sawir gelinaya
- Structured output
- Shay JSON
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $1.42 / $0.355 / $6.07
- Aan la faragelin
MiMo-V2.5-Pro
W8A16Xiaomi · 1.02T MoE · 42B active · 8-bit
Model pro ah oo yar oo loogu talagalay wax-soo-saar habaysan oo sax ah.
Id-ga model-kaMiMo-V2.5-Pro-W8A16- Context
- 256K
- Input
- Qoraal gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $0.50 / $0.125 / $2.00
- Aan la faragelin
MiMo-V2.5
W8A16Xiaomi · 8-bit multimodal · W8A16
Model yar oo degdeg ah oo loogu talagalay hawlo fudud qiime aad u yar.
Id-ga model-kaMiMo-V2.5-W8A16- Context
- 256K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $0.50 / $0.125 / $2.00
- Aan la faragelin
Hy3
W8A16Tencent · 295B MoE · 21B active · 8-bit
Model miisaaniyad hoose oo loogu talagalay jawaabo wada-hadal oo kooban.
Id-ga model-kaHy3-W8A16- Context
- 256K
- Input
- Qoraal gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Reasoning effort
- Prompt caching
- Streaming
- Gudaha / cached / dibadda halkii 1M
- $0.50 / $0.125 / $2.00
Model-yada Shannon hal qiime go'an input-ka iyo output-ka
-
Aan la faragelinShannon 1.6 Lite
Jawaabaha degdega ah, hufan ee hawlaha maalinlaha ah
Id-ga model-kashannon-1.6-lite- Context
- 192K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Raadinta webka
- Streaming
- Qiimaha / 1M token
- $3.90
-
Aan la faragelinShannon 1.6 Pro
Sababta sare ee dhibaatooyinka adag
Id-ga model-kashannon-1.6-pro- Context
- 192K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Raadinta webka
- Streaming
- Qiimaha / 1M token
- $7.80
-
Aan la faragelinShannon 2 Lite
Shannon 2 Lite: pipeline degdeg ah oo fikir-qoris-dib u eegis ah oo leh raad reasoning la arki karo. Chat maalinle ah, soo koobid iyo Q&A.
Id-ga model-kashannon-2-lite- Context
- 192K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Raadinta webka
- Streaming
- Qiimaha / 1M token
- $3.90
-
Aan la faragelinShannon 2 Pro
Shannon 2 Pro: marxalado fikir-qoris-dib u eegis oo ka qoto dheer falanqayn, qoraal dheer iyo reasoning tallaabo-badan.
Id-ga model-kashannon-2-pro- Context
- 192K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Raadinta webka
- Streaming
- Qiimaha / 1M token
- $5.85
-
Aan la faragelinShannon 3
Model Shannon oo jiilka ugu dambeeya ah oo leh reasoning degdeg ah qiime hoose oo go'an.
Id-ga model-kashannon-3- Context
- 192K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Raadinta webka
- Streaming
- Qiimaha / 1M token
- $3.35
-
Aan la faragelinShannon 3 Pro
Heer Shannon 3 oo awood sare leh oo loogu talagalay shaqo adag oo tallaabo-badan.
Id-ga model-kashannon-3-pro- Context
- 192K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Raadinta webka
- Streaming
- Qiimaha / 1M token
- $3.35
-
Aan la faragelinShannon 3.1
Shannon 3 oo ku shaqeeya qalabkayaga, streaming xawaare buuxa ah iyada oo aan lahayn xad rate ah.
Id-ga model-kashannon-3.1- Context
- 192K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Raadinta webka
- Streaming
- Qiimaha / 1M token
- $3.35
-
Aan la faragelinShannon 3.1 Pro
Reasoning qoto dheer oo xawaare buuxa ah oo leh marxalad cilmi-baaris gadaal ah oo loogu talagalay su'aalaha adag.
Id-ga model-kashannon-3.1-pro- Context
- 192K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Raadinta webka
- Streaming
- Qiimaha / 1M token
- $3.35
-
Aan la faragelinShannon 3.5
One streaming answer that thinks, searches, draws charts and diagrams and writes files inline.
Id-ga model-kashannon-3.5- Context
- 192K
- Input
- Qoraal + sawir gelinaya
- Structured output
- JSON schema
- Tools
- Streaming
- Qiimaha / 1M token
- $3.35
-
Aan la faragelinShannon Coder 1
Khabiir code iyo tools: wax-ka-beddelka wakiilka, faylal dhaadheer, iyo wareegyo tool oo adag.
Id-ga model-kashannon-coder-1- Context
- 128K
- Input
- Qoraal gelinaya
- Structured output
- JSON schema
- Tools
- Reasoning
- Streaming
- Qiimaha / 1M token
- $8.00
Daalacasho dukumentiyada
- Bilaw degdeg ah Samee fure, SDK-gaaga ku jiheey Shannon oo dir codsigii ugu horreeyay.
- Model-yo & qiimo Id kasta oo model ah oo leh context window, input iyo qiimaha halkii 1M token.
- Xaqiijinta Hal fure endpoint kasta: header-ka Bearer ama x-api-key.
- Streaming Server-sent events oo leh reasoning deltas; usage wuxuu imaanayaa chunk-ga ugu dambeeya.
- Wacyigelinta Shaqada Qeex tools, u ogolow model-ka inuu wacdo, hagi isagoo adeegsanaya tool_choice.
- Waxsoosaarka qaabaysan Jawaabo sida shay JSON ah ama schema-gaaga JSON.
- Effort-ka reasoning-ka Akhri qulqulka reasoning-ka; deji effort-ka model-yada open-weight ee la martigeliyo.
- Kaydinta Prompt-ka Model-yada la martigeliyo si automatic ah ayay u cache-gareeyaan prefix-yada prompt-ka ee soo noqnoqda, 25% qiimaha input-ka.
- Responses API Lahjadda OpenAI Responses ee isla model-yadan iyo isla furahan.
- Qaabka Anthropic Hayso SDK-gaaga Anthropic iyo code-ka Messages; beddel base URL-ka.
- Qalabka CLI ee coding-ka Claude Code iyo Codex CLI iyadoo Shannon yahay model-ka ka dambeeya.
- Khaladaadka Maareynta Koodhka xaaladda, noocyada khaladka iyo qaabka jidhka khaladka.
Diyaar ma u tahay inaad dhisto?
Heer bilaash ah ayaa ku jira akoon kasta. Furahaagu waa hal guji u jirta.