kalmantic/PeakInfer

Analyze LLM inference points in your code for cost, latency, throughput, and reliability issues

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
pathPath to analyzeno./src
peakinfer-tokenPeakInfer API token (required). Get your token at https://peakinfer.com/dashboard. Pass via secrets.PEAKINFER_TOKEN. yes
github-tokenGitHub token for PR comments. Defaults to github.token if not provided.no
runtime-sourceRuntime data source: helicone, langfuseno
runtime-api-keyAPI key for runtime sourceno
events-filePath to runtime events file (JSONL)no
events-mapField mappings for non-standard formatsno
include-benchmarksCompare to InferenceMAX benchmarksnotrue
benchmark-frameworkBenchmark framework: api, vllm, tginoapi
evals-sourceEvals source: braintrust, evalicano
evals-api-keyAPI key for evals sourceno
fail-on-criticalFail the action if critical issues are foundnofalse
comment-modeWhen to post comments: always, on-issues, nevernoalways
show-enhancement-promptsShow prompts for adding more layersnotrue
namedescription
verdictPASS, OK, REVIEW, BLOCK, PAUSED, SKIP, ERROR
critical-countNumber of critical issues
warning-countNumber of warnings
drift-countNumber of drift detections
layers-usedComma-separated list of active layers
credits-usedCredits consumed (paid mode)
credits-remainingCredits remaining (paid mode)
inference-pointsNumber of inference points found