Feat/gemma3 model intelligence #487
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Context :
This is a follow-up enhancement to my previous PR (the core MCP Bridge). It adds a layer of system intelligence to ensure users run the version of Gemma 3 best suited for their specific hardware.
Description :
This PR introduces a Hardware Intelligence Layer to the Gemma 3 MCP Gateway. While Gemma 3 offers a wide range of model sizes (1B to 27B), users often face "out of memory" errors or poor performance by selecting a model that exceeds their local hardware capabilities.
This update allows the gateway to "think" about the system it is running on, automatically recommending and defaulting to the optimal Gemma 3 variant based on detected system RAM.
Key Changes
system_utils.py: Added a diagnostic utility using psutil to detect system RAM and map it to recommended Gemma 3 variants (1B, 4B, 12B, or 27B).
config.py: Centralized model mapping and introduced support for the GEMMA_MODEL_SIZE environment variable for flexible deployment.
server.py: Integrated a "Startup Intelligence Report" that logs detected hardware and the selected model choice to the console on boot.