Skip to main content

LLMs made simple

OpenAI-compatible API access, guided tools, and a clean model catalog on Cloudflare's edge.

Get a key, open the dashboard, and ship faster without platform noise.

FREECreate an account and open your dashboard in minutes

Start free with Google and open your dashboard in seconds.

Open dashboard

LLM.kiwi gives you an OpenAI-compatible API, web tools, and a transparent model catalog on Cloudflare's global network. Sign in with Google or GitHub to manage keys and start building.

Read our Privacy Policy · Terms of Service · Cookie Policy

For AI BeginnersFor API DevelopersFor Businesses
AI Beginners
Guided tools and workflows so anyone can start shipping useful AI output today.

Start with beginner-friendly tools, clear prompts, and fast results.

Open beginner path
API Access
Drop in to the OpenAI-compatible API, use auto routing, and deploy quickly.
curl
curl -X POST \\
  "https://api.llm.kiwi/v1/chat/completions" \\
  -H "Authorization: Bearer sk_kiwi_..." \\
  -H "Content-Type: application/json" \\
  -d '{"model":"auto","messages":[{"role":"user","content":"Hello from the edge."}]}'
API docsModels catalog
Business Integrations
Launch chatbots, internal copilots, and workflow automations for real teams.

Use LLM.kiwi for support bots, lead qualification, and operations assistants.

Open business page
Edge Infrastructure

Delivered on Cloudflare's global edge network

Lower-latency routing and one consistent runtime path keep production rollouts simpler.

  • Near-user routing for faster first-token response.
  • Fewer avoidable cross-region hops.
  • One runtime path for API traffic and guided tools.
Cloudflare logoGlobal edge inference network
Explore edge deployment
Cloudflare edge network visualization
Why teams trust LLM.kiwi

Clarity where most AI products add noise

Secure keys, guided tools, clear model choices, and edge delivery—without the extra clutter.

Developer onboarding
A compact integration surface
One endpoint, one key pattern, and one sensible default model.
  • OpenAI-compatible endpoint
  • Clear model defaults
  • Dashboard-driven key provisioning
View developer path
Popular
Guided workflows
AI tools that start with structure
Outlines, rewrites, newsletters, and summaries without a blank-page start.
  • Purpose-built web tools
  • Default-key-powered usage
  • Fewer blank-page decisions
Explore guided workflows
Operational fit
Edge delivery with local-language coverage
Use auto for general traffic and switch to Croatian-first workflows when fit matters.
  • Cloudflare edge delivery
  • Croatian-first model support
  • Straightforward path from evaluation to rollout
Learn more

Specialized workflow collections

Pick a focused toolkit for faster practical output.

View all audience pages
AI Tools for Bloggers
Refresh posts, generate outlines, and ship SEO-ready content faster.
Explore blogging tools
AI Tools for Content Creators
Create scripts, repurpose content, and scale multi-channel publishing.
Explore creator tools
AI Tools for Marketers
Build campaign copy, ad variants, and conversion messaging quickly.
Explore marketing tools
Full Stack AI Apps

Launch full-stack apps without boilerplate

Start from deploy-ready app patterns for real LLM.kiwi integrations.

Starter architecture patternsEnvironment setup guidesDeploy-ready references
Browse starter appsIntegration docs

Platform benefits

Ship faster without giving up reliability or model clarity.

OpenAI Drop-In
Keep your existing SDK calls while switching routing, policies, and model lanes behind one gateway.
Security by Default
API keys are hashed at rest, revocable instantly, and managed with account-level defaults for controlled internal usage.
Model Clarity
Public callers use the auto route while engineering retains full model lineage visibility and deployment control.
Model Catalogue

Find the right model in seconds

Browse model lineage, exact upstream names, and usage notes in one place. For public API calls, use auto.

Owner + model mappingCapability filtersIntegration-ready IDs
Explore modelsAPI reference
Simple Pricing

Free to Start, Pro to Scale

Pro unlocks 5x resources, direct advanced models, coding agents, enterprise rate limiting, and unlimited image generation for production workloads.

Free
€0.00
Perfect for testing, learning, and first deployments.
  • All core dashboard features plus AUTO as the recommended default endpoint model and direct hrLLM access.
  • Starter API keys, tools, and playground sessions.
  • Baseline rate limiting suitable for normal daily usage.
Create free accountRead API docs
Pro
€29.99/month
Or €200.00/year
Save €159.88 yearly (€16.67/month effective)
  • 5x resource limits across API, playground, and tools.
  • Direct access to qwen3, kimik2, and gpt-oss-120b.
  • Enterprise rate limiting and higher sustained throughput.
  • Unlimited image generation plus advanced agent tools.
  • Automated tier enforcement, expiry handling, and secure webhook sync.
Upgrade monthlyUpgrade yearly
Use AI Responsibly

Committed to responsible AI

We publish our Responsible AI policy and take part in the Responsible AI audit program.

Ready to build?

Start with one account and one key

Create an account, provision a key, and start building. No credit card required.

Use the official Google button or One Tap to create your free account.

Open dashboard

Free tier includes API access, AI Tools Web, the AUTO endpoint, and hrLLM for Croatian-first workflows.