Update deployment guide

README.md

@@ -22,7 +22,7 @@ library_name: chronos-forecasting
 
 # Chronos-2
 
-**Update Dec 30, 2025:** ☁️ Deploy Chronos-2 on Amazon SageMaker. [New guide](https://github.com/amazon-science/chronos-forecasting/blob/main/notebooks/deploy-chronos-to-amazon-sagemaker.ipynb) covers real-time GPU and CPU inference, serverless endpoints (run on demand, no idle costs), and batch transform for large-scale forecasting.
+**Update Jun 5, 2026:** ☁️ Deploy Chronos-2 on AWS with AutoGluon-Cloud. Real-time, serverless, or batch inference in 3 lines of code — pandas DataFrames in, forecasts out. Check out the [new deployment guide](https://auto.gluon.ai/cloud/stable/tutorials/foundation-model-timeseries.html).
 
 **Chronos-2** is a 120M-parameter, encoder-only time series foundation model for zero-shot forecasting.
 It supports **univariate**, **multivariate**, and **covariate-informed** tasks within a single architecture.
@@ -31,7 +31,8 @@ Trained on a combination of real-world and large-scale synthetic datasets, it ac
 Chronos-2 is also **highly efficient**, delivering over 300 time series forecasts per second on a single A10G GPU and supporting both **GPU and CPU inference**.
 
 ## Links
-- 🚀 [Deploy Chronos-2 on Amazon SageMaker](https://github.com/amazon-science/chronos-forecasting/blob/main/notebooks/deploy-chronos-to-amazon-sagemaker.ipynb)
+- ☁️ [Deploy on SageMaker with AutoGluon-Cloud](https://auto.gluon.ai/cloud/stable/tutorials/foundation-model-timeseries.html) (recommended)
+- 🚀 [Deploy on SageMaker with JumpStart](https://github.com/amazon-science/chronos-forecasting/blob/main/notebooks/deploy-chronos-to-amazon-sagemaker.ipynb)
 - 📄 [Technical report](https://arxiv.org/abs/2510.15821v1)
 - 💻 [GitHub](https://github.com/amazon-science/chronos-forecasting)
 - 📘 [Example notebook](https://github.com/amazon-science/chronos-forecasting/blob/main/notebooks/chronos-2-quickstart.ipynb)
@@ -53,9 +54,7 @@ Chronos-2 is also **highly efficient**, delivering over 300 time series forecast
 🧩 Chronos & Chronos-Bolt do not natively support future covariates, but they can be combined with external covariate regressors (see [AutoGluon tutorial](https://auto.gluon.ai/1.4.0/tutorials/timeseries/forecasting-chronos.html#incorporating-the-covariates)). This only models per-timestep effects, not effects across time. In contrast, Chronos-2 supports all covariate types natively.
 
 
-## Usage
-
-### Local usage
+## Running the model locally
 
 For experimentation and local inference, you can use the [inference package](https://github.com/amazon-science/chronos-forecasting).
 
@@ -76,8 +75,7 @@ pipeline = Chronos2Pipeline.from_pretrained("amazon/chronos-2", device_map="cuda
 context_df = pd.read_parquet("https://autogluon.s3.amazonaws.com/datasets/timeseries/electricity_price/train.parquet")
 
 # (Optional) Load future values of covariates
-test_df = pd.read_parquet("https://autogluon.s3.amazonaws.com/datasets/timeseries/electricity_price/test.parquet")
-future_df = test_df.drop(columns="target")
+future_df = pd.read_parquet("https://autogluon.s3.amazonaws.com/datasets/timeseries/electricity_price/test.parquet").drop(columns="target")
 
 # Generate predictions with covariates
 pred_df = pipeline.predict_df(
@@ -91,14 +89,44 @@ pred_df = pipeline.predict_df(
 )
 ```
 
-### Deploying a Chronos-2 endpoint to SageMaker
+## Production use on Amazon SageMaker
+
+For production use, we recommend deploying Chronos-2 to Amazon SageMaker. There are two options:
+
+- **AutoGluon-Cloud** (recommended) — minimal setup with a high-level Python API: pass a pandas DataFrame in, get forecasts back. Supports real-time, serverless, and batch inference out of the box.
+- **SageMaker JumpStart** — fine-grained control over the deployment configuration. JSON request/response payloads only; serverless inference and batch prediction require additional setup.
 
-For production use, we recommend deploying Chronos-2 endpoints to Amazon SageMaker.
+### ☁️ AutoGluon-Cloud
+
+Install AutoGluon-Cloud:
+
+```
+pip install autogluon.cloud>=0.5.0
+```
+
+Make predictions from a pandas DataFrame
+
+```python
+from autogluon.cloud import TimeSeriesFoundationModel
+
+model = TimeSeriesFoundationModel(model_name="chronos-2")
+
+# Batch prediction
+forecast_df = model.predict(df, prediction_length=24)
+
+# Deploy & invoke a real-time endpoint
+endpoint = model.deploy(instance_type="ml.g5.xlarge")
+forecast_df = endpoint.predict(df, prediction_length=24)
+```
+
+For more details (e.g. serverless endpoints, covariate-aware forecasting), see the [full deployment guide](https://auto.gluon.ai/cloud/stable/tutorials/foundation-model-timeseries.html).
+
+### 🚀 SageMaker JumpStart
 
 First, update the SageMaker SDK to make sure that all the latest models are available.
 
 ```
-pip install -U sagemaker
+pip install -U 'sagemaker<3'
 ```
 
 Deploy an inference endpoint to SageMaker.
@@ -116,21 +144,18 @@ predictor = model.deploy()
 Now you can send time series data to the endpoint in JSON format.
 
 ```python
-import pandas as pd
-df = pd.read_csv("https://raw.githubusercontent.com/AileenNielsen/TimeSeriesAnalysisWithPython/master/data/AirPassengers.csv")
-
 payload = {
     "inputs": [
-        {"target": df["#Passengers"].tolist()}
+        {"target": [1.0, 2.5, ..., 12.3]}
     ],
     "parameters": {
-        "prediction_length": 12,
+        "prediction_length": 24,
     }
 }
 forecast = predictor.predict(payload)["predictions"]
 ```
 
-For more details about the endpoint API, check out the [example notebook](https://github.com/amazon-science/chronos-forecasting/blob/main/notebooks/deploy-chronos-to-amazon-sagemaker.ipynb)
+For more details about the endpoint API, check out the [example notebook](https://github.com/amazon-science/chronos-forecasting/blob/2.2.2/notebooks/deploy-chronos-to-amazon-sagemaker.ipynb).
 
 
 ## Training data