v3.2.0 · Stable release

Build
Faster.

Powerful tools for modern developers — clean APIs, SDKs in 8 languages, and documentation that actually makes sense.

npm
pip
go
gem
$ npm install devlaunch-sdk
quickstart.js
JS
Python
Go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
import DevLaunch from 'devlaunch-sdk'; // Initialize with your API key const client = new DevLaunch({ apiKey: process.env.DEVLAUNCH_KEY, region: 'us-east-1', }); // Send your first request const response = await client.process({ input: 'Hello, DevLaunch', model: 'v3-turbo', stream: false, }); console.log(response.output); // → { id: "req_8fK2p...", result: "...", tokens: 42 } // That's it. 3 lines to your first result.
JavaScript · Node 18+ ✓ No errors
Documentation

Docs worth reading.

Written for developers, not product managers. Every page has a working code example.

Getting Started
Quick Start
Authentication
SDKs & Libraries
Core API
Requests
Responses
Streaming
New
Webhooks
Advanced
Rate Limits
Error Handling
Batch Requests
New
Reference
API Reference
Changelog
Migration Guide
Docs / Getting Started / Quick Start
Quick Start
Updated: Nov 2024 Read time: 4 min v3.2.0
Get your first API response in under 2 minutes. Install the SDK, set your API key, and send a request. That's the whole setup.
Install the SDK
Terminal
$ npm install devlaunch-sdk
Initialize the client
JavaScript
import DevLaunch from 'devlaunch-sdk';

const client = new DevLaunch({ apiKey: 'dl_live_...' });
Parameters
apiKey
string
Your DevLaunch API key. Required. Never expose this client-side — use environment variables.
region
string
Processing region. Defaults to nearest region. Options: us-east-1, eu-west-1, ap-southeast-1.
timeout
number
Request timeout in milliseconds. Default: 30000. Set lower for latency-sensitive applications.
retries
number
Auto-retry count on transient errors (5xx). Default: 2. Set to 0 to disable.
APIs

Clean endpoints. Consistent design.

REST API with OpenAPI 3.1 spec. Every response is predictable. Every error is useful.

POST
/v3/process
Submit a processing request
200
GET
/v3/requests/{id}
Get request status and result
200
GET
/v3/requests
List requests with pagination
200
DEL
/v3/requests/{id}
Cancel a pending request
204
POST
/v3/batch
Submit up to 100 requests in one call
202
GET
/v3/usage
Get token and request usage stats
200
PUT
/v3/webhooks/{id}
Update webhook endpoint config
200
GET
/v2/models ·deprecated
Migrate to /v3/models
301
POST /v3/process
200 OK
Request
Response
cURL
{
  "model": "v3-turbo",
  "input": "Summarize this document...",
  "options": {
    "max_tokens": 512,
    "temperature": 0.7,
    "stream": false
  },
  "metadata": {
    "user_id": "usr_abc123"
  }
}
Examples

Copy. Paste. Ship.

Working examples in every language we support. All tested against the live API.

Basic Request
JavaScript
Streaming Response
JavaScript
Batch Processing
Python
Webhook Handler
Node.js
Error Handling
TypeScript
basic-request.js
JavaScript
1
2
3
4
5
6
7
8
9
10
11
12
13
import DevLaunch from 'devlaunch-sdk'; const client = new DevLaunch({ apiKey: process.env.DL_KEY }); const result = await client.process({ model: 'v3-turbo', input: 'Analyze the sentiment of this text', options: { max_tokens: 256 }, }); console.log(result.output); console.log(`Tokens used: ${result.usage.total_tokens}`);
Output →
{ sentiment: "positive", confidence: 0.94, tokens: 38 }
124ms
streaming.js
JavaScript
1
2
3
4
5
6
7
8
9
10
11
12
const stream = await client.stream({ model: 'v3-turbo', input: 'Write a short story about the sea', }); for await (const chunk of stream) { process.stdout.write(chunk.delta); } // Tokens counted as they stream console.log('\nDone.', stream.usage);
Streaming →
The sea does not forgive
live
batch.py
Python
1
2
3
4
5
6
7
8
9
10
11
12
from devlaunch import DevLaunch client = DevLaunch(api_key="dl_live_...") batch = client.batch.submit( requests=[ {"input": item, "model": "v3-fast"} for item in my_documents ] ) print(f"Batch {batch.id}: {batch.total} requests queued")
Output →
Batch bat_9xKp2: 84 requests queued
0.3s
webhook.js
Node.js
1
2
3
4
5
6
7
8
9
10
11
12
13
app.post('/webhook', (req, res) => { const sig = req.headers['dl-signature']; const event = client.webhooks.verify({ payload: req.body, signature: sig, secret: process.env.DL_WEBHOOK_SECRET, }); if (event.type === 'request.completed') { handleResult(event.data); } res.json({ received: true });
Verified →
event.type: "request.completed" · HMAC-SHA256 ✓
2ms
error-handling.ts
TypeScript
1
2
3
4
5
6
7
8
9
10
11
12
13
14
import { DevLaunch, DLError, RateLimitError } from 'devlaunch-sdk'; try { const result = await client.process({ ... }); } catch (err) { if (err instanceof RateLimitError) { // Wait and retry await sleep(err.retryAfter); } else if (err instanceof DLError) { console.error(err.code, err.message); // err.code is always a string, never undefined } }
Caught →
RateLimitError: retry_after=2400ms · code="rate_limit_exceeded"
typed
Community

Built with 40,000 developers.

An active ecosystem — open source, real support, and people who've solved the problem you're hitting right now.

🐙
GitHub
The SDK is fully open source. Browse the code, file issues, submit PRs, or fork it for your own use. We ship fast and we ship in public.
4,200 stars · 218 contributors
Star the repo →
💬
Discord
18,000 developers in the server. Share what you're building, get help debugging, or just lurk in #showcase and see what's possible.
18,400 members · 240 active today
Join the server →
📚
Forum
Long-form Q&A that doesn't disappear. Search before you ask — the answer is probably already there. Every question gets a response.
12,800 threads · avg 4hr response
Browse threads →
Community Activity
Live
MS
marcos_s merged a PR — "Add retry jitter to rate limit handler"
PR
2m ago
AK
akaur.dev opened an issue — "Streaming drops last token on long responses"
Issue
8m ago
JL
j_liu started a discussion — "How are people handling webhook retries at scale?"
Disc
15m ago
NR
nadia_r merged a PR — "Python SDK: add async context manager support"
PR
34m ago
TK
t_kowalski opened an issue — "Go SDK: type assertion panics on null metadata field"
Issue
1h ago
Pricing

Free until you scale.

Every plan includes full API access. Upgrade when your usage demands it — no feature gating on the free tier.

Free
$
0
/ month · forever
API Quota
500 requests / month
Build, test, and prototype with no time limit and no credit card. Full API access, no hidden restrictions.
  • 500 API requests / month
  • All models (rate limited)
  • SDKs in all 8 languages
  • Community support
  • Public API reference
Get Free API Key
Enterprise
Custom
Volume pricing · dedicated SLA
API Quota
Unlimited · dedicated infra
For high-volume production systems with compliance requirements, dedicated infrastructure, and contractual SLAs.
  • Unlimited API requests
  • Dedicated infrastructure
  • SOC 2 & HIPAA compliance
  • 99.99% SLA + credits
  • Dedicated solutions engineer
  • Custom rate limits
Talk to Sales
Get Started

Your API key is
one command away.

Free forever on the Free tier. No credit card. No expiry. Just build.

Free tier · 500 req/month · All languages · No card required
terminal
$ npm install devlaunch-sdk added 1 package in 1.3s $ devlaunch init ? Paste your API key: dl_live_●●●●●●●●●●● ✓ API key verified ✓ Config saved to .devlaunch.env $ devlaunch test ✓ Connection OK · us-east-1 · 18ms ✓ You're ready to build. $