Infron announced that Qwen3.8 27B is now available for free on its inference platform. The model is served as qwen/qwen3.8-27b:free and billed at $0 per token — free-model requests never deduct from your credit balance.
- Model:
qwen/qwen3.8-27b:free— a 27B multimodal model with a 256K-token context window (text, image, and video input), flexible thinking that can be enabled or disabled, streaming, function calling, vision, and JSON mode. - Pricing: $0 in and out, with zero data retention.
- Free tier rules: Standard free models on Infron require an account balance of $5 or more (held, not spent, as an anti-abuse measure) and share a pool of 1,000 requests per day per account, resetting at 00:00 UTC.
- How to access: Select the free model at infron.ai or call it through the OpenAI-compatible endpoint at
llm.onerouter.pro.