Qwen3.8 27B is free on 13 providers today. Permanent free tiers come first, then one-time trials.
| Provider | API model ID | Context | Free limits | Sign-up |
|---|---|---|---|---|
| Groq | qwen/qwen3.8-27b |
131K | 30 RPM; 8K TPM on most chat models (varies by model) | No card |
| Cloudflare Workers AI | @cf/qwen/qwen3.8-27b |
262K | Varies by model | No card |
| OVH AI Endpoints | Qwen3.8-27B |
Long context | 2 RPM per IP per model (anonymous, free); 400 RPM with a key (pay-as-you-go) | Free account |
| ModelScope | Qwen/Qwen3.8-27B |
— | 500 requests/day per model | Phone |
| Moark (Gitee AI) | qwen3.8-27b |
— | Not published | Phone |
| LLMTR | qwen/qwen3.8-27b-free |
131K | Not published; lower allowance in an account's first 24h | Free account |
| Hetzner Inference API | Qwen3.8-27B |
262K | 10 requests, 4M input / 100K output tokens per 60s (per API key) | Card required |
| Regolo.ai Trial | qwen3.8-27b |
— | Not published (trial) | No card |
| VLM Run Gateway Trial | qwen/qwen3.8-27b |
168K | Anonymous: 10/min, 30/hr per IP; with free key: 240/min, 10,000/hr | No card |
| Hyperbolic Trial | qwen3.8-27b |
— | 300 requests/minute per API key | Phone |
| Novita AI Trial | qwen/qwen3.8-27b |
Long context | 60 RPM | Free account |
| Scaleway Generative APIs Trial | qwen3.8-27b |
Long context | Organization quota in tokens/minute (see console) | Free account |
| IteraCompute Trial | qwen/qwen3.8-27b |
128K | 160K tokens/minute (fair use, may queue); 128K context, 8K output | Free account |
Free lists change often. Each provider page shows its full limits and code examples, and every change is logged in the changelog. Want to call it from Claude Code or Cursor? Use the setup generator.