galaktikuzay/galaktikuzay-llm-eval

Galaktik Uzay LLM Eval

A static public demo for the MIT-licensed multilingual LLM evaluation toolkit maintained by Galaktik Uzay.

The toolkit evaluates OpenAI-compatible endpoints across:

• —JSON validity and schema compliance
• —required-field validation
• —source URL adherence
• —Turkish and multilingual language consistency
• —latency, reliability, retries and response-format fallbacks
• —token usage and estimated cost
• —JSON, CSV and Markdown reporting

This Space uses illustrative sample data only. It does not call model APIs, request API keys, or process production data.

Source

• —Toolkit: https://github.com/ozguradmin/galaktikuzay/tree/master/llm-eval
• —Main repository: https://github.com/ozguradmin/galaktikuzay
• —Live platform: https://galaktikuzay.com