FastAPIframework~8 mins

Multiple response types in FastAPI - Performance & Optimization

Choose your learning style9 modes available

Learn Why Deep Visual Try Challenge Project Recall Perf

Performance: Multiple response types

MEDIUM IMPACT

This affects server response time and client rendering speed by controlling how data is serialized and sent over the network.

Returning different response formats based on client needs

FastAPI

from fastapi import FastAPI
from fastapi.responses import JSONResponse
from fastapi.templating import Jinja2Templates
from fastapi import Request

app = FastAPI()
templates = Jinja2Templates(directory="templates")

@app.get("/item")
async def get_item(request: Request, response_type: str = "json"):
    data = {"item": "value"}
    if response_type == "html":
        return templates.TemplateResponse("item.html", {"request": request, "data": data})
    return JSONResponse(content=data)

Using template rendering for HTML and JSONResponse for JSON ensures efficient serialization and smaller, optimized payloads.

📈 Performance GainReduces CPU overhead and payload size, improving LCP by ~50ms and lowering server load

Returning different response formats based on client needs

FastAPI

from fastapi import FastAPI
from fastapi.responses import JSONResponse, HTMLResponse

app = FastAPI()

@app.get("/item")
async def get_item(response_type: str):
    data = {"item": "value"}
    if response_type == "html":
        return HTMLResponse(content=f"<html><body>{data['item']}</body></html>")
    else:
        return JSONResponse(content=data)

Manually constructing HTML strings and switching response types in the endpoint can cause inconsistent serialization and larger payloads.

📉 Performance CostAdds extra CPU for string concatenation and larger HTML payload increases LCP by ~50ms on slow networks

Performance Comparison

Pattern	DOM Operations	Reflows	Paint Cost	Verdict
Manual HTML string + JSON switch	Minimal	Minimal	Higher due to larger payload	[X] Bad
Template rendering + JSONResponse	Minimal	Minimal	Lower due to optimized payload	[OK] Good

Rendering Pipeline

The server serializes data into the chosen response type, sends it over the network, and the browser parses and renders it accordingly.

→Server Serialization

→Network Transfer

→Browser Parsing

→Rendering

⚠️ BottleneckServer Serialization and Network Transfer

Core Web Vital Affected

LCP

This affects server response time and client rendering speed by controlling how data is serialized and sent over the network.

Optimization Tips

1Use FastAPI's built-in JSONResponse and template rendering for different response types.

2Avoid manual string concatenation for HTML to reduce CPU and payload size.

3Choose response types that minimize payload size to improve LCP.

Performance Quiz - 3 Questions

Test your performance knowledge

Which response type choice most improves Largest Contentful Paint (LCP) in FastAPI?

AManually building HTML strings for all responses

BUsing JSONResponse for JSON and template rendering for HTML

CReturning plain text for all responses

DSending large HTML payloads regardless of client needs

DevTools: Network

How to check: Open DevTools, go to Network tab, reload the endpoint, and inspect the response size and content type.

What to look for: Look for smaller payload size and correct content-type headers indicating efficient response format.