(doc) add and refine fields to model_collection.yml

[leohearts]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello @leohearts, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request focuses on significantly enhancing the model_collection.yml GitHub issue template. The primary objective is to enable the collection of more detailed and structured information about AI models by introducing new fields and refining existing ones, thereby improving the accuracy and completeness of model documentation.

Highlights

• Enhanced Model Information Capture: I've added several new input fields to the model_collection.yml issue template, including 'Model Family', 'Model URL', 'Release Date', 'Context Window (Tokens)', 'Parameters (Billions)', 'Training Data Cutoff', and 'License'. These additions aim to gather more comprehensive and structured data about AI models.
• Structured Architecture Data: A new dropdown field for 'Architecture' has been introduced, offering predefined options like 'Transformer', 'Mixture of Experts (MoE)', and 'Other'. This will help standardize the classification of model architectures.
• Refined Existing Fields and Options: Descriptions for 'Provider', 'Model Endpoints', and 'Additional Information' have been updated for improved clarity and guidance. Additionally, 'tool-call' was corrected to 'tool-calling' under 'Model Capabilities', and 'json' was added as a new option for 'Model Modalities'.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in issue comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist is currently in preview and may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments to provide feedback.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request aims to improve the issue template for collecting new model information by adding several new fields and refining existing ones. The descriptions and placeholders are much clearer, which is a great improvement for usability.

However, there are some critical issues. The main problem is that all the new fields being added (like Model Family, Model URL, Release Date, etc.) are not reflected in the backend Go Model struct. This means the data collected via this form cannot be processed by the existing automation, rendering these additions ineffective.

Additionally, there are a couple of data inconsistencies:

• The capability tool-calling is used, while the backend expects tool-call.
• A json output modality is added, which is not a recognized type in the system.

These issues need to be addressed to ensure the collected data is accurate and usable. My detailed comments are on the relevant lines in the diff.

[gemini-code-assist]

This PR introduces the Model Family field, and several others later in the file (e.g., Model URL, Release Date, Context Window). However, these new fields are not defined in the backend Model struct in cmd/tools/jem-generator/types/types.go.

As a result, any data provided for these new fields in an issue will not be processed by the jem-generator and will be lost. This makes the additions non-functional.

To fix this, the Model struct in cmd/tools/jem-generator/types/types.go and the generator logic that uses it must be updated to handle these new fields. Was this intended to be part of a different PR? Without the backend changes, this PR is incomplete.

[gemini-code-assist]

The capability tool-calling is inconsistent with the value expected by the backend. The generator code in cmd/tools/jem-generator/types/types.go defines this capability as tool-call. Using a different value here will lead to data processing issues. Please use the value defined in the system to ensure consistency.

        - label: "tool-call"

[gemini-code-assist]

The output modality json is not a recognized value in the backend. The Modality type in cmd/tools/jem-generator/types/types.go only defines text, audio, image, video, and vector. Submitting json will cause data inconsistencies. Please either remove this option or update the backend to support it.

[LemonNekoGH]

Thanks for your contribution! I will review this after my #5