fix: sync batch size with sample size in gradio app

fix: Increase timeout

feat: Add batch size configuration in api and frontends

fix: Add rate limiter to Kisski connector

docs: Add evalutation results

feat: Support more providers

feat: Add security against malicious clients

fix: Fix config files

docs: Add instructions on how to setup nginx, certbot and systemd configuration

feat: Add registry to support any kind of inference provider

docs: Add link to HF demo

docs: Add clarifying text to gradio app

perf: cache blbl schema at module load instead of rebuilding per request

docs/deps: update HF deployment guide, pin gradio>=6.9, document HF_TOKEN scope requirement

cleanup: remove spaces import and @spaces.GPU decorators (hardware changed to cpu-basic)

fix: increase @spaces.GPU timeout to 300s to avoid GPU task abort on slow LLM calls

fix: catch exceptions in do_evaluate to show error in UI instead of crashing ZeroGPU runtime

fix: add @spaces.GPU decorator to satisfy ZeroGPU spaces check; graceful fallback when spaces not installed

fix: explicitly set hardware: cpu-basic in Space metadata to suppress spaces.GPU check

fix: disable SSR mode in Gradio launch to prevent Node.js server crash on HF Spaces

fix: remove local package install from requirements.txt (HF Spaces copies source directly)

fix: replace editable install (-e .) with plain . in requirements.txt for HF Spaces compatibility

Fix HF yaml

feat: Add Gradio app for HF Spaces deployment

docs: Add documentation on webservices and HF container

feat: Add webservice for demonstration

Add report for whole-dataset performance experiment

Add timeout parameter

Support batched requests

Fix schema, add docs on iterative workflow

Revert split-author entries in gold and adapt global instructions

Add support for global rules

Add `--grep` and skill to iteratively improve TEI element descriptions

Add support for <surname> and <forename>

Add disclaimer & license

Improve identifcation of `<label>`

Add human-readable output for comparison of gold and llm-annotation

Add human-inspectable to evaluate-llm.py output via `--show-annotations`

Fix issue with `&` character

Improve Gliner model loading

Rename function to be more generic

Add detailed explanation

Add evaluation

Update .gitignore

full implementation

Update implementation plan

First commit