Merge pull request #15 from IBM/Bedrock_using_litellm

Bedrock application using litellm

Author: adharshctr

Bedrock-using-litellm/.python-version

@@ -0,0 +1 @@
+3.13

Bedrock-using-litellm/README.md

@@ -0,0 +1,62 @@
+This guide explains how to set up and run your **Python app (`main.py`)** using **`uv`** for environment and dependency management.
+
+## Prerequisites
+
+Make sure the new system has:
+- **Python 3.8+**
+- **Internet access**
+- **`uv`** installed
+
+If `uv` is not installed, run one of the following commands:
+
+```bash
+curl -LsSf https://astral.sh/uv/install.sh | sh
+```
+or
+```bash
+pip install uv
+```
+## Steps for running the application
+
+### 1. Sync Dependencies
+
+Install all dependencies listed in `pyproject.toml` and `uv.lock`:
+
+```bash
+uv sync
+```
+
+### 2. Activate the Environment
+
+You can manually activate the environment:
+
+```bash
+source .venv/bin/activate
+```
+
+### 3. Update the envs in `.env` file
+
+Update **AWS Bedrock** creds such as 
+- AWS_ACCESS_KEY_ID
+- AWS_SECRET_ACCESS_KEY
+- AWS_SESSION_TOKEN
+
+
+Refer following documentation to know about **Traceloop environmental variables**:
+https://www.ibm.com/docs/en/instana-observability/1.0.300?topic=started-traceloop-instrumentation#configuring-the-environment
+
+
+
+### 4. For the collection of metrics, Install OTel Data Collector for GenAI (ODCG)
+
+To install the data collector, Refer: https://www.ibm.com/docs/en/instana-observability/1.0.300?topic=started-install-otel-data-collector-genai-odcg
+
+### 5. Run the App
+
+Run the application using:
+
+```bash
+uv run main.py
+```
+### Output
+![Traces](image.png)

Bedrock-using-litellm/image.png

@@ -0,0 +1,21 @@
+from litellm import completion
+import litellm
+litellm.callbacks = ["otel"]
+
+from dotenv import load_dotenv
+load_dotenv()
+
+from traceloop.sdk import Traceloop
+Traceloop.init(app_name="LitellmApp")
+
+# Example: Using Amazon Titan Text Express on Bedrock
+response = completion(
+    model="bedrock/amazon.titan-text-express-v1",
+    messages=[
+        {"role": "system", "content": "You are an assistant that summarizes text."},
+        {"role": "user", "content": "Summarize this: Large language models are transforming the way software is built."},
+    ],
+)
+
+print("Response from Bedrock model:")
+print(response['choices'][0]['message']['content'])

Bedrock-using-litellm/main.py

@@ -0,0 +1,11 @@
+[project]
+name = "bedrock-using-litellm"
+version = "0.1.0"
+description = "Add your description here"
+readme = "README.md"
+requires-python = ">=3.13"
+dependencies = [
+    "litellm[proxy]>=1.79.0",
+    "python-dotenv>=1.2.1",
+    "traceloop-sdk==0.47.4",
+]