Self-hosted LLM deployment needed?
├─ Yes, need maximum throughput → vLLM
├─ Yes, need absolute max GPU efficiency → TensorRT-LLM
├─ Yes, local development only → Ollama
└─ No, use managed API (OpenAI, Anthropic) → No serving layer needed

undefined

**Key Parameters:**
- `--dtype`: Model precision (auto, float16, bfloat16)
- `--max-model-len`: Context window size
- `--gpu-memory-utilization`: GPU memory fraction (0.8-0.95)
- `--tensor-parallel-size`: Number of GPUs for model parallelism

from fastapi import FastAPI
from fastapi.responses import StreamingResponse
from openai import OpenAI
import json

app = FastAPI()
client = OpenAI(base_url="http://localhost:8000/v1", api_key="not-needed")

@app.post("/chat/stream")
async def chat_stream(message: str):
    async def generate():
        stream = client.chat.completions.create(
            model="meta-llama/Llama-3.1-8B-Instruct",
            messages=[{"role": "user", "content": message}],
            stream=True,
            max_tokens=512
        )

        for chunk in stream:
            if chunk.choices[0].delta.content:
                token = chunk.choices[0].delta.content
                yield f"data: {json.dumps({'token': token})}\n\n"

        yield f"data: {json.dumps({'done': True})}\n\n"

    return StreamingResponse(
        generate(),
        media_type="text/event-stream",
        headers={"Cache-Control": "no-cache"}
    )

// Integration with ai-chat skill
const sendMessage = async (message: string) => {
  const response = await fetch('/chat/stream', {
    method: 'POST',
    headers: { 'Content-Type': 'application/json' },
    body: JSON.stringify({ message })
  })

  const reader = response.body!.getReader()
  const decoder = new TextDecoder()

  while (true) {
    const { done, value } = await reader.read()
    if (done) break

    const chunk = decoder.decode(value)
    const lines = chunk.split('\n\n')

    for (const line of lines) {
      if (line.startsWith('data: ')) {
        const data = JSON.parse(line.slice(6))
        if (data.token) {
          setResponse(prev => prev + data.token)
        }
      }
    }
  }
}

import bentoml
from bentoml.io import JSON
import numpy as np

@bentoml.service(
    resources={"cpu": "2", "memory": "4Gi"},
    traffic={"timeout": 10}
)
class IrisClassifier:
    model_ref = bentoml.models.get("iris_classifier:latest")

    def __init__(self):
        self.model = bentoml.sklearn.load_model(self.model_ref)

    @bentoml.api(batchable=True, max_batch_size=32)
    def classify(self, features: list[dict]) -> list[str]:
        X = np.array([[f['sepal_length'], f['sepal_width'],
                       f['petal_length'], f['petal_width']] for f in features])
        predictions = self.model.predict(X)
        return ['setosa', 'versicolor', 'virginica'][predictions]

from langchain_openai import OpenAIEmbeddings, ChatOpenAI
from langchain_community.vectorstores import Qdrant
from langchain.chains import RetrievalQA
from langchain.text_splitter import RecursiveCharacterTextSplitter

undefined

GPU Memory (GB) = Model Parameters (B) × Precision (bytes) × 1.2

undefined

undefined

@bentoml.api(
    batchable=True,
    max_batch_size=32,
    max_latency_ms=1000  # Wait max 1s to fill batch
)
def predict(self, inputs: list[np.ndarray]) -> list[float]:
    # BentoML automatically batches requests
    return self.model.predict(np.array(inputs))

services:
  - name: vllm-service
    url: http://vllm-llama-8b:8000
    plugins:
      - name: rate-limiting
        config:
          minute: 60  # 60 requests per minute per API key
      - name: key-auth
      - name: prometheus

from prometheus_client import Counter, Histogram

requests_total = Counter('llm_requests_total', 'Total requests')
tokens_generated = Counter('llm_tokens_generated', 'Total tokens')
request_duration = Histogram('llm_request_duration_seconds', 'Request duration')

@app.post("/chat")
async def chat(request):
    requests_total.inc()
    start = time.time()
    response = await generate(request)
    tokens_generated.inc(len(response.tokens))
    request_duration.observe(time.time() - start)
    return response

Frontend (React) → API Gateway → vLLM Server → GPU Inference
     ↑                                                  ↓
     └─────────── SSE Stream (tokens) ─────────────────┘

User Query → LangChain
              ├─> Vector DB (Qdrant) for retrieval
              ├─> Combine context + query
              └─> LLM (vLLM) for generation

Client → API → Message Queue (Celery) → Workers (vLLM) → Results DB

python scripts/benchmark_inference.py \
  --endpoint http://localhost:8000/v1/chat/completions \
  --model meta-llama/Llama-3.1-8B-Instruct \
  --concurrency 32 \
  --requests 1000

undefined

undefined

MODEL_ROUTING = {
    "small": "meta-llama/Llama-3.1-8B-Instruct",  # Fast, cheap
    "large": "meta-llama/Llama-3.1-70B-Instruct", # Accurate, expensive
    "code": "codellama/CodeLlama-34b-Instruct"    # Code-specific
}

@app.post("/chat")
async def chat(message: str, task: str = "small"):
    model = MODEL_ROUTING[task]
    # Route to appropriate vLLM instance

import tiktoken

def estimate_cost(text: str, model: str, price_per_1k: float):
    encoding = tiktoken.encoding_for_model(model)
    tokens = len(encoding.encode(text))
    return (tokens / 1000) * price_per_1k

undefined