As recorded at collection: Kluster.ai offers up to 1,000 free batch requests per file
As recorded at collection: The API service Kluster.ai, which sends multiple requests to open-source AI models at once, supports up to 1,000 requests per file and files up to 100MB under its free usage allowance. Signup and payment requirements, as well as how to activate the free allowance, need to be confirmed.
- What you get
- Up to 1,000 free batch requests per file · Maximum file size 100MB
- Price
- Batch inference for small models from $0.10 per 1 million tokens · 72-hour completion window
- Valid until
- To be confirmed
- Requirements
- Up to 1,000 requests per file · Maximum file size 100MB · Signup and payment requirements need to be confirmed
Details
Kluster.ai is an AI cloud platform for running open-source AI models such as DeepSeek-R1, Llama 4 Maverick/Scout, and Qwen3. Its batch processing handles multiple requests at once, with the free allowance supporting up to 1,000 requests per file and files up to 100MB. Promotional credits are also offered periodically, but their amounts and distribution dates are not specified.
It provides an OpenAI-compatible API (api.kluster.ai/v1) that works with the OpenAI Python SDK. Real-time inference is also supported, with response latency stated to be under 1 second. Signup instructions and the steps to activate the free allowance need to be confirmed.
Batch inference for small models starts at $0.10 per 1 million tokens, with a 72-hour completion window. The promotion end date and signup and payment requirements need to be confirmed.
Sources and verification history
- Collection channel
- aicredits
- Collected at
- October 1, 2026, 04:21 KST
- First published
- Updated
- Main post
- View the main post