ContextQMD
Back to Camel

Using real OpenAI

components/camel-ai/camel-langchain4j-agent/test-execution.md

latest1.4 KB
Original Source

Test Execution

macOS or Linux Without an NVIDIA Graphics Card

If Ollama is already installed on the system, execute the tests with:

bash
mvn verify -Dollama.endpoint=http://localhost:11434/ -Dollama.model=granite4:3b -Dollama.instance.type=remote

The Ollama Docker image is very slow on a MacBook without NVIDIA hardware acceleration.

Linux With an NVIDIA Graphics Card

Hardware acceleration can be used, and the tests can be executed with:

bash
mvn verify -Dollama.container.enable.gpu=enabled

OpenAI or OpenAI-Compatible Endpoints

To run tests against OpenAI or any OpenAI-compatible endpoint (including local Ollama via its OpenAI-compatible API):

bash
# Using real OpenAI
mvn verify -Dollama.instance.type=openai \
    -Dopenai.api.key=sk-your-api-key \
    -Dopenai.model=gpt-4o-mini

# Using local Ollama as OpenAI-compatible endpoint
mvn verify -Dollama.instance.type=openai \
    -Dopenai.endpoint=http://localhost:11434/v1/ \
    -Dopenai.model=granite4:3b \
    -Dopenai.api.key=dummy

Note that LangChain4jAgentWrappedFileIT and LangChain4jAgentMultimodalityIT require multimodal models.

All tests (tools and multimodal) can be executed with an OpenAI instance and the model qwen3-vl:2b-instruct:

bash
mvn verify -Dollama.instance.type=openai \
    -Dopenai.endpoint=http://localhost:11434/v1/ \
    -Dopenai.model=qwen3-vl:2b-instruct \
    -Dopenai.api.key=dummy