For teams running GPTCache, LangChain, or a self-built Redis cache

Your semantic cache is probably wrong more often than your dashboards show

A plain similarity threshold can't tell a true match from a near-miss in the gray zone — in our own tests against real support traffic, precision there was under 2%. CacheVerifier adds a synchronous verification call plus fine-tuning on your own feedback to close that gap, without you having to switch cache backends.

integrations/gptcache_adapter.py — drop into an existing GPTCache pipeline
from integrations.gptcache_adapter import CacheVerifierEvaluation

cache.init(
    similarity_evaluation=CacheVerifierEvaluation(api_key="cv_..."),
    ...
)

Implements GPTCache's own SimilarityEvaluation extension point — no fork, no PR required. Already using LangChain or a self-built Redis lookup? /v1/verify is a plain REST call.

Synchronous verification

POST /v1/verify takes a (query, candidate_answer) pair and returns approve/reject with a score and latency — call it inline on any gray-zone hit, whatever cache backend produced it.

Self-serve fine-tuning

Fine-tuning is cheap and fast (seconds to minutes) so it's free to run yourself: POST /v1/feedback accumulates your own correctness signal, POST /v1/finetune/jobs trains a tenant-specific verifier on it.

Drift & stability monitoring

GET /v1/monitor/drift-status watches your gray-zone positive rate for the kind of drift that quietly turns a fine-tuned verifier harmful — no extra inference required.

Not just claimed — checked

We tried to break this before shipping it

Three questions the underlying research didn't answer — how noisy feedback can get, how little data is enough, whether gains decay over time — checked on three public benchmarks and real production support traffic, not just one dataset.

Fine-tuning gain (AUC delta) vs. label noise — all three datasets cross zero in the same range

-0.13-0.06+0.00+0.07+0.130%5%10%20%30%40%LmArenaSearchQueriesQuora

4

datasets, including real support traffic

~30%

label-noise threshold, consistent across datasets

500–1000

feedback rows before fine-tuning reliably helps

How it works

01

Start free

Register gets you a tenant, a dashboard, and an API key — no card required.

02

Point your gray zone at /v1/verify

Wherever your cache backend flags an uncertain hit — GPTCache, LangChain, a self-built Redis lookup.

03

Run a health check

See baseline vs. fine-tuned precision on a sample of your real traffic before committing to anything.

04

Let us run the loop

Move to managed calibration + drift monitoring once you know it's worth automating.

Everything is free during beta.

Verify and self-serve fine-tuning stay free forever. The health check report and managed calibration are free for now too, in exchange for your feedback while we validate the product.

Start verifying