Skip to content

Conversation

manju956
Copy link
Collaborator

  • Contains separate Containerfiles for HMC server and Client
  • logic to parse HMC related configurations and pass to HMC server and client
  • Modifications in HMC server to run on different port
  • Modifications in HMC client to accept OPENAI_BASE_URL and MCP_SERVER_URL at runtime

@manju956 manju956 self-assigned this Jul 30, 2025
@manju956 manju956 added the enhancement New feature or request label Jul 30, 2025
Copy link
Member

@dharaneeshvrd dharaneeshvrd left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Add a README under examples/hmc-agent, explaining the expected configs need to be set by the user.
Also instructions on how to build
Add a README inside app also what's what and steps to build server and agent containers.

@@ -54,7 +55,10 @@ async def interactive_chat(self):

async def main():
try:
client = MCPClient()
model = os.getenv("OLLAMA_MODEL")
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Its not OLLAMA_MODEL right?
extract it from llm.conf

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this is model param from llmArgs, will rename it

Comment on lines 19 to 20
var_to_add=MCP_SERVER_URL=$(jq -r '.mcpServerURL' /etc/pim/pim_config.json)
sed -i "/^MCP_SERVER_URL=.*/d" /etc/pim/hmc.conf && echo "$var_to_add" >> /etc/pim/hmc.conf
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Are you expecting this so that user need to set it directly via ip instead of localhost:8003/sse?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I am expecting user to provide MCP server URL with PIM partition's IP eg: http://9.114.98.3:8003/sse

@manju956
Copy link
Collaborator Author

Add a README under examples/hmc-agent, explaining the expected configs need to be set by the user. Also instructions on how to build Add a README inside app also what's what and steps to build server and agent containers.

Addressed documentation. ptal


## Configuration

Since vLLM AI image can be used as a base for many LLM inferencing applications like chatbot, entity extraction and many more. Below provided configurations tune the vLLM engine as per the AI use case.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
Since vLLM AI image can be used as a base for many LLM inferencing applications like chatbot, entity extraction and many more. Below provided configurations tune the vLLM engine as per the AI use case.
Since vLLM AI image can be used as a base for many LLM inferencing applications like chatbot, entity extraction and many more. Below provided configurations tune the vLLM engine as per the HMC Agent AI use case.

@dharaneeshvrd
Copy link
Member

How much is the final image size Mac?

@adarshagrawal38
Copy link
Member

How much is the final image size Mac?

It is around 1.6 GB

@dharaneeshvrd
Copy link
Member

Cool, how about after installing pyarrow and streamlit?

@adarshagrawal38
Copy link
Member

Cool, how about after installing pyarrow and streamlit?

Streamlit installation was falling yesterday, will take the updated Container file and try again

@dharaneeshvrd
Copy link
Member

@manju956 please squash the commits before merging

@manju956 manju956 merged commit b71959c into main Aug 5, 2025
1 check passed
@dharaneeshvrd dharaneeshvrd deleted the hmc_agent branch August 29, 2025 05:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants