[go: up one dir, main page]

Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update to use the platform standard model name / version in samples and snippets and imports (for Google AI and Vertex AI) #460

Closed
rachelsaunders opened this issue Jun 24, 2024 · 1 comment
Assignees
Labels
feature New feature or request

Comments

@rachelsaunders
Copy link
rachelsaunders commented Jun 24, 2024

For both Google AI and Vertex AI, we try to use the most capable and developer-friendly model as possible in our samples, snippets, etc.

The Gemini 1.5 models can understand both text-only and multimodal (even in chat) inputs, which relieves the developer from switching b/w Gemini 1.0 Pro and Gemini 1.0 Pro Vision. For both platforms, there's also a stable version of each Gemini 1.5 model available.

For the foreseeable future, we have the guidance of using Gemini 1.5 Flash for all samples and snippets... unless the feature/capability isn't supported on Gemini 1.5 Flash. In those cases, use Gemini 1.5 Pro (which supports more features, but has a lower RPM).

This is the guidance for the explicit model names in code:

  • For Google AI: use gemini-1.5-flash (or gemini-1.5-pro when required)
  • For Vertex AI: use gemini-1.5-flash-001 (or gemini-1.5-pro-001 when required)
@kevinthecheung
Copy link
Contributor

I have addressed this to the extent that in the docs, Gemini 1.5 Flash is the canonical model example, and Gemini 1.5 Pro is the canonical example of a model that is not Flash.

I have not used the names of specific versions (gemini-1.5-flash-001) because I believe that would circumvent the nice defaults provided by Genkit.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
feature New feature or request
Projects
None yet
Development

No branches or pull requests

3 participants