IBM Watsonx
JD

Scale AI workloads
with confidence.

An enterprise-ready data and AI platform designed to accelerate the impact of AI across your business with transparency, governance, and speed.

3x

Time to value

Faster deployment

40%

Cost reduction

Lower infrastructure spend

99.9%

Trust score

Enterprise governance

180+

Global reach

Countries supported

The foundation for
modern AI.

Generative AI

Build, train, and tune foundation models with your own data.

Data Store

A fit-for-purpose data store built for AI workloads.

SELECT * FROM data_lake LIMIT 100

Governance

Monitor and manage models to ensure responsibility and compliance.

Compliant

Hybrid Cloud

Run anywhere: on-prem, public cloud, or edge.

Automation

Orchestrate workflows and automate business processes.

Analytics

Turn data into insights with powerful BI tools.

Developer first.

Access powerful APIs, SDKs, and pre-built models. Integrate seamlessly into your existing DevOps pipelines. Built on open technologies like Red Hat OpenShift.

  • Python, Java, and Node.js SDKs
  • Zero-copy data integration
  • Enterprise-grade security controls
model_train.pyconfig.jsondeploy.yaml
from watsonx import Model, Data

# Initialize model
model = Model(
    model_id="ibm/granite-13b-chat-v2",
    credentials={
        "apikey": "YOUR_API_KEY",
        "url": "https://us-south.ml.cloud.ibm.com"
    }
)

# Prepare fine-tuning data
dataset = Data.load("enterprise_docs_v2.json")

# Start training job
job = model.train(
    data=dataset,
    epochs=10,
    learning_rate=2e-5
)

print(f"Training job started: {job.id}")