Akkio API
Agent-ready OpenAPI document · curated by JenticAI/MLMl InferenceapiKey16 EndpointsREST
Know of an official OpenAPI document? Contribute it →
For Agents
Upload datasets, train predictive models, run predictions, and ask natural-language data questions on Akkio without managing ML infrastructure.
Use for: I need to train a predictive model on a CSV dataset without writing ML code, Make a prediction using a trained Akkio model from inside an agent workflow, Upload a new dataset to Akkio for analysis, Check the status of a running Akkio model training task
Not supported: Does not handle deep learning model training, real-time streaming inference, or LLM fine-tuning - use for no-code tabular predictive modeling and Chat Explore queries only.
Jentic publishes the only available OpenAPI specification for Akkio API, keeping it validated and agent-ready. Akkio is a no-code predictive AI platform that lets teams upload tabular datasets, train classification and regression models, and run predictions through a REST API. The same API exposes a Chat Explore endpoint for asking natural-language questions about a dataset, plus project and dataset management for organising data across teams. Authentication uses an API key supplied either as an X-API-Key header or an api_key query parameter.
Install Jentic One Beta
Connect the Akkio API to your agent
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Akkio API, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
1
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fakkio.com%2Fakkio" | sh2
Step 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fakkio.com%2Fakkio" | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
Capabilities
What an agent can do with Akkio API.
Upload a tabular dataset to Akkio with appended rows or replace an existing dataset by name
Submit a model training task on a chosen dataset and target column, then poll for status and results
Make a prediction by sending input rows to a trained Akkio model
Run a Chat Explore query against a dataset and retrieve the conversational answer when ready
Organise work across teams by creating, updating, and deleting Akkio projects
List all datasets and models in an account to wire predictions into downstream agent workflows
Use Cases
Patterns agents use Akkio API for, with concrete tasks.
★ No-code churn prediction for SaaS teams
A customer success team uploads a dataset of historical accounts and churn outcomes to Akkio, trains a classification model on the churned column, and then calls the prediction endpoint nightly with active accounts to score churn risk. Because Akkio handles feature processing and training internally, the team avoids managing notebooks or ML pipelines. End-to-end setup takes hours rather than weeks.
POST /api/v1/models/train/new with the customers dataset and target column 'churned', poll /api/v1/models/train/{task_id}/status until complete, then POST /v1/models with active account rows to get churn probabilities
Conversational analysis of a sales dataset
An analyst loads a quarterly sales dataset into Akkio and uses Chat Explore to ask questions like 'which region grew the most last quarter?' through the API. The /api/v1/chat-explore/new endpoint accepts the question, returns a task ID, and the result endpoint delivers a structured answer once the model finishes. This embeds natural-language data Q&A into internal tools without exposing the underlying tabular data.
POST /api/v1/chat-explore/new with the sales dataset id and the question, then GET /api/v1/chat-explore/result/{task_id} once the status endpoint reports completion
Lead scoring inside a CRM workflow
A sales operations team trains an Akkio regression model on closed-won deals and exposes the prediction endpoint to a CRM automation. As new leads enter the CRM, the agent posts the lead's attributes to the Akkio model and writes back a lead score on the record. Akkio handles the training and inference; the CRM stays the system of record.
POST /v1/models with model_id and an input row containing lead attributes, then update the matching CRM record with the predicted score
Agent-driven dataset and model lifecycle
A Jentic agent receives an intent like 'train a model on this dataset', uploads the rows via POST /v1/datasets, submits a training task, polls for completion, and finally calls the prediction endpoint with new inputs. The same agent can list existing models, retire stale ones, and create new Akkio projects for separate use cases. The full loop runs without a human opening the Akkio UI.
Search Jentic for 'train a predictive model on a dataset', load the Akkio training operation schema, execute POST /api/v1/models/train/new and chain the status and result calls to deliver a trained model id back to the user
Key Endpoints
16 endpoints — jentic publishes the only available openapi specification for akkio api, keeping it validated and agent-ready.
METHOD
PATH
DESCRIPTION
POST
/v1/datasets
Create a dataset or append rows
POST
/api/v1/models/train/new
Submit a model training task
GET
/api/v1/models/train/{task_id}/status
Get training task status
POST
/v1/models
Make a prediction with a trained model
POST
/api/v1/chat-explore/new
Submit a Chat Explore natural-language query
GET
/api/v1/chat-explore/result/{task_id}
Retrieve a Chat Explore result
POST
/api/v1/projects
Create a new project
POST
/v1/datasets
Create a dataset or append rows
POST
/api/v1/models/train/new
Submit a model training task
GET
/api/v1/models/train/{task_id}/status
Get training task status
POST
/v1/models
Make a prediction with a trained model
POST
/api/v1/chat-explore/new
Submit a Chat Explore natural-language query
GET
/api/v1/chat-explore/result/{task_id}
Retrieve a Chat Explore result
POST
/api/v1/projects
Create a new project
Why Jentic?
What agents get from Jentic-routed access to this vendor.
Setup
Wiring Akkio by hand means sending its X-API-Key, launching training or chat-explore jobs and polling their task status, and writing your own retry logic. Through Jentic you install once, import Akkio from the API Directory, store the key once, and your agent calls it.
Permission scoping
Akkio carries its dataset and model details in the request body and reads jobs back by task_id, so scope your agent by operation: limit it to the operations it needs, such as training a model or running a Chat Explore query, and leave out project creation it does not use.
Credential isolation
Your Akkio key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'train a predictive model on a dataset', and Jentic returns the matching Akkio operation with its input schema so the agent calls the right endpoint without browsing the reference docs.
Related APIs
Alternatives and complements available in the Jentic catalogue.
FAQs
Specific to using Akkio API through Jentic.
Why is there no official OpenAPI spec for Akkio API?
Akkio does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call Akkio API via structured tooling. It is validated against the live API and kept up to date. Get started with Jentic One, the self-hosted execution layer.
What authentication does the Akkio API use?
Akkio uses an API key passed either as the X-API-Key request header or as an api_key query parameter. Jentic stores the key encrypted in its credential vault and injects it at execution time, so the raw key never enters the agent's prompt or response context.
Can I train a predictive model on a CSV dataset with the Akkio API?
Yes. POST your rows to /v1/datasets to register the dataset, then POST /api/v1/models/train/new with the dataset reference and target column. Poll /api/v1/models/train/{task_id}/status until the task completes, then call /api/v1/models/train/{task_id}/result for the trained model id.
What are the rate limits for the Akkio API?
The OpenAPI spec does not declare per-endpoint rate limits. Plan-based limits apply on Akkio's side; check your account tier in the Akkio dashboard before running large batch prediction or training jobs.
How do I run a prediction with the Akkio API through Jentic?
Search Jentic for 'make a prediction with an Akkio model', load the operation schema for POST /v1/models, and execute it with the model id and an input rows array. Jentic returns the prediction payload directly to the agent.
Does Akkio support natural-language questions over a dataset?
Yes, via Chat Explore. POST /api/v1/chat-explore/new with the dataset id and your question to get a task id, then GET /api/v1/chat-explore/result/{task_id} once the status endpoint reports completion to retrieve the answer.
Can I limit what my agent is allowed to do with the Akkio API?
Yes. Because you run Jentic One yourself, your own rules decide which Akkio operations and credentials your agent may use. Akkio carries its dataset and model details in the request body and reads jobs back by task_id, so you scope the agent by operation: allow only the calls it needs, such as POST /api/v1/models/train/new to train a model or POST /api/v1/chat-explore/new to run a Chat Explore query, and leave out operations like project creation that it does not use. The stored API key is injected only for the operations you permit, so the agent cannot reach endpoints you have not granted.
GET STARTED