The Sovereign AI Compute & MCP Factory

import tensorfact as tf

# Initialize Sovereign Client with MCP
client = tf.Client(api_key="tf_live_sec_89f02a...")

# Connect Model Context Protocol tool servers
client.mcp.connect_server("postgres_db", uri="mcp://db.internal:8080")
client.mcp.connect_server("sandbox", uri="mcp://vm.internal:9000")

# Stream inference with tool execution
response = client.chat.create(
    model="deepseek-ai/DeepSeek-V3",
    messages=[{
        "role": "user", 
        "content": "Audit Q3 ledger anomalies"
    }],
    stream=True,
    temperature=0.2
)

for chunk in response:
    print(chunk.delta.content, end="", flush=True)

import { TensorFact } from '@tensorfact/sdk';

const tf = new TensorFact({
  apiKey: process.env.TENSORFACT_API_KEY
});

// Stream completions via WebSockets
const stream = await tf.chat.completions.create({
  model: 'meta-llama/Llama-3.3-70B-Instruct',
  messages: [{
    role: 'user', 
    content: 'Execute ERP reconciliation'
  }],
  mcp_tools: ['erp_gateway', 'inventory_db'],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content || '');
}

curl https://api.tensorfact.com/v1/chat/completions \
  -H "Authorization: Bearer $TENSORFACT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-ai/DeepSeek-V3",
    "messages": [
      {"role": "user", "content": "Analyze yield"}
    ],
    "mcp_enabled": true,
    "stream": true
  }'

use tensorfact::{Client, ChatOptions};

let client = Client::new("tf_live_key...");
let mut stream = client.stream_chat(ChatOptions {
    model: "qwen/Qwen2.5-72B-Instruct".into(),
    prompt: "Optimize robotics vector".into(),
    ..Default::default()
}).await?;

while let Some(token) = stream.next().await {
    print!("{}", token?);
}