Cloudflare Workers AI: 60+ Free AI Models via 9Router
July 10, 2026 · 10 min read
Some links below are affiliate links. I may earn a commission at no extra cost to you.
What Is Cloudflare Workers AI?
Cloudflare Workers AI gives you access to 60+ AI models through Cloudflare's global network. It includes LLMs, image generation models, embedding models, and more. Best of all: the free tier gives you 10,000 requests per day.
Getting Access
You need a Cloudflare account and API token. The token needs Workers AI permissions. Create one in the Cloudflare dashboard under API Tokens.
Connecting Through 9Router
9Router already supports Cloudflare AI as a provider. Just add your API key in the 9Router dashboard and use the cf/ prefix for models:
curl http://localhost:20128/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{"model":"cf/@cf/qwen/qwen2.5-coder-32b-instruct","messages":[{"role":"user","content":"Hello!"}]}' Best Models Available
| Model | Response Time | Best For |
|---|---|---|
| Qwen 2.5 Coder 32B | ~8s | Coding, technical tasks |
| Llama 3.3 70B | ~6s | General purpose, reasoning |
| DeepSeek R1 Distill 32B | ~15s | Complex reasoning |
| Mistral Small 3.1 24B | ~5s | Fast responses, multilingual |
| Llama 3.2 3B | ~2s | Quick tasks, classification |
Rate Limits
- Free tier: 10,000 requests/day across all models
- Rate limit: 1,000 requests per 10 minutes
- Concurrent: Up to 6 simultaneous requests
For most personal projects, the free tier is more than enough. I've never hit the limit even with heavy daily use.
Use Cases
- Chat bots — Free LLM inference for customer support bots
- Code review — Automatic code analysis with Qwen Coder
- Content generation — Blog posts, summaries, translations
- Image generation — Via Cloudflare's Stable Diffusion models
- Embeddings — Vector embeddings for RAG systems
This article contains affiliate links. I may earn a commission at no extra cost to you.