Kaimei attaches to the models you already run and tells you why each one decides what it decides. Then it watches them in production, so drift, bugs and regressions surface with their reasons attached.
pip install kaimei
Free plan included
POST https://api.kaimei.ai/v1/explanations
Why Kaimei
Two plausible models can agree on an answer and believe completely different things. In the open Titanic example, a random forest and a logistic regression both say Mr Braund died. One blames his third-class ticket. The other says no ticket would have saved him. Kaimei makes those reasons visible for every model you ship, so you can tell which one you can defend.
How it works
The client reads the live model in your notebook. Everything it learns also appears in the dashboard, for the people who never open a notebook.
One call reads the fitted estimator in your process and uploads its layers and weights. Kaimei never stores the original pickle, .h5 or .pt file.
client.models.upload(clf,
name="Churn v4",
X=X_ref, y=y_ref)
Ask why for any input. KernelSHAP, permutation Shapley or LIME for tabular models; Grad-CAM, occlusion and LIME for images.
client.explanations.create(
model="churn-v4",
input=row, method="shap")
Study the model as a whole: importance, learned rules, behaviour curves and interactions. Insights land in the dashboard as they finish.
client.analyses.create(
model="churn-v4",
type="full")
Log what your stack already predicts, run shadow models, and catch drift with the explanation that names the culprit.
client.predictions.create(
model="churn-v4",
input=row, output=out)
Platform
Local attributions, counterfactuals, global analysis, versioning and production monitoring, in one place and under one API.
For a model Kaimei runs itself with up to 12 inputs, KernelSHAP enumerates every feature coalition over a fixed reference sample. The answer is exact for that sample, and the same request returns the same numbers.
Every layer, weight and bias of an uploaded structure is drawn, and a trace walks a prediction through it. Repeated blocks fold, so a 110M-parameter encoder stays readable.
The smallest change that flips the decision, step by step, with the features you choose held fixed.
Log predictions, run shadow models and measure drift per input. A fare arriving in pence shows up as significant drift within a single batch.
Ship v2 as a new version and compare its insights with v1. Same accuracy, different reasons, is a result you can finally see.
If Kaimei cannot run a model itself, it can still explain and monitor it. Point it at a scoring endpoint, send attributions you compute yourself, or upload shapes only and keep the weights at home.
inference="remote"POST /v1/explanationsweights="summary"Grad-CAM, occlusion and LIME on photos, drawn exactly as the dashboard draws them. Every heatmap comes with named regions, a focus region holding half the evidence, and a check of what the model says with that region hidden.
Dashboard
Each project shows traffic, explained share and open alerts across its models. Open a model to edit any feature and re-explain, read the insights an analysis produced, or walk the trees.
Developers
The Python client uploads a fitted estimator in one call and turns explanations into sentences and charts. Every call is also a plain HTTP request, with typed errors and automatic retries.
from kaimei import Kaimei
client = Kaimei(api_key="km_live_...")
explanation = client.explanations.create(
model="credit-risk-v3",
prediction_id="pred_8f2a19",
input={"income": 72000, "credit_score": 681, "utilization": 0.41},
method="shap",
)
print(explanation.output) # 0.81 decline
print(explanation.top_features) # credit_score -0.22from kaimei import Kaimei
from kaimei.tabular import plot_insight
client = Kaimei() # KAIMEI_API_KEY from the environment
model = client.models.upload(
clf, # scikit-learn, Keras/TensorFlow, PyTorch
name="Churn v4",
feature_names=FEATURES,
X=X_train, y=y_train, # reference rows explanations compare against
)
analysis = client.analyses.create(model=model.slug, type="full")
client.analyses.wait(analysis.id)
for insight in client.insights.list(model=model.slug).data:
plot_insight(insight, model=model)curl https://api.kaimei.ai/v1/explanations \
-H "Authorization: Bearer km_live_..." \
-H "Content-Type: application/json" \
-d '{
"model": "credit-risk-v3",
"prediction_id": "pred_8f2a19",
"input": {"income": 72000, "credit_score": 681, "utilization": 0.41},
"method": "shap"
}'Built for teams who answer for their models
Where the model is small enough, Kaimei enumerates every feature coalition rather than sampling. Deterministic requests mean the number in the report is the number in the API.
Kaimei never stores your original model file. Upload shapes only and the weights stay in your environment. Point it at your own endpoint and the model never leaves at all.
The Python client is Apache 2.0. Every call is plain REST. Every example on this site runs end to end in a public notebook you can open in Colab.
Start on the free plan: 10 models, 500 explanations, 20 analyses and 5,000 logged predictions a month. No sales call required.