Awan LLM

Awan LLM is an inference API that charges a flat monthly fee instead of per token, with limits set on requests rather than tokens. It targets developers running agents, roleplay apps, code completion or bulk data processing who want predictable costs.

The company runs its own datacenters and GPUs, which is how it offers unlimited tokens up to each model's context length. Models such as Meta Llama 3.1 8B and 70B are available, and requests are not filtered for content.

Each tier sets requests per minute and per day for small and medium models, with parallel requests from the Pro plan. The service says it keeps no logs of prompts or outputs. Payments are taken through PayPal only.