Qwen/Qwen3.8-27BQwen3.8 27B is an LLM listed in RunInfra Model APIs. RunInfra serves it as Qwen/Qwen3.8-27B at $0.00 per 1M input and output tokens until Aug 18, 2026, 11:00 AM UTC, with standard rates resuming automatically. Its context window is 262,144 tokens. The API provides OpenAI-compatible chat completions.
USD, pay per token
Input and output tokens are listed at $0.00 until .
Standard rates resume automatically: $0.10 per 1M input tokens, $0.40 per 1M output tokens.
Hosted inference is available to every account. You pay for token usage from the same workspace balance.
Output speed
Time to first token
Confirm how your client reaches this model.
Check the limits your workload must fit.
See which request modes the API supports.
Set RUNINFRA_GATEWAY_KEY to your workspace API key before using an example.
Verify the company and operating credentials behind this API.
RunInfra is a sub-product of RightNow Research Lab.
Y CombinatorBacked by Y Combinator.
NVIDIA InceptionMember of NVIDIA Inception.
RunInfraby RightNow© 2026 RunInfra. All rights reserved.
