Current collected state Current
Observed since 20 Oct, 19:04 UTC
No superseded metadata states collected. Unchanged observations do not create new revisions.
adapterci
Regression CI for fine-tuned local models. Scores travel, data doesn't.
Regression CI for fine-tuned local models. Runs entirely against your own OpenAI-compatible endpoint (Ollama, LM Studio, llama.cpp, vLLM). Only metric names and floats are ever sent to the AdapterCI API — no prompts, no completions, no weights.
pip install adapterci
adapterci init # writes adapterci.yaml
# edit adapterci.yaml: api_key, project, endpoint, baseline_model, candidate_model
adapterci # runs the suite, pushes scores, exits 1 on regression
Add domain-specific test cases in a local JSONL file (never uploaded):
{"prompt": "Summarize in one sentence: ...", "check": "contains", "expected": "..."}
check can be contains, regex, numeric_answer, or json_schema_keys.
export ADAPTERCI_API_BASE=http://localhost:8787
MIT
Download counts and popularity will appear here as the collector receives them.
| Package / target | Requirement | Relationship |
|---|---|---|
httpxEnvironment{
"extras": "",
"marker": null,
"requires_python": ">=3.10",
"evaluation": "NOT_EVALUATED"
} | httpx>=0.27 | runtime |
pyyamlEnvironment{
"extras": "",
"marker": null,
"requires_python": ">=3.10",
"evaluation": "NOT_EVALUATED"
} | pyyaml>=6.0 | runtime |
bdist_wheel · 4.68 KB · Python py3
sdist · 3.88 KB · Python source
Observed since 20 Oct, 19:04 UTC
No superseded metadata states collected. Unchanged observations do not create new revisions.
package_version:c4c4b7a8c3d654e2797ec6a45bad555634df2c200e7f3fddc4e3298d1337bf6b