Bump Gemini models to 2.5 #5285
Draft
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
Contributor
There was a problem hiding this comment.
Pull request overview
This PR updates Gemini model references from version 2.0 to version 2.5 across the codebase. The changes include updating model names from preview versions (e.g., gemini-2.5-flash-preview-04-17) to stable versions (e.g., gemini-2.5-flash), and bumping all 2.0 models to their 2.5 equivalents.
Key Changes:
- Updated Gemini model versions from 2.0 to 2.5 across all test configurations
- Changed from preview model names to stable release names (removed
-preview-04-17suffix) - Updated documentation and examples to reference the new model versions
Reviewed changes
Copilot reviewed 25 out of 26 changed files in this pull request and generated 1 comment.
Show a summary per file
| File | Description |
|---|---|
| ui/fixtures/config/tensorzero.toml | Updated variant model references from preview to stable 2.5-flash |
| tensorzero-optimizers/tests/common/gcp_vertex_gemini_sft.rs | Updated optimizer test model from 2.0-flash-lite-001 to 2.5-flash-lite |
| tensorzero-core/tests/e2e/providers/google_ai_studio_gemini.rs | Updated all test provider model names from 2.0 to 2.5 versions |
| tensorzero-core/tests/e2e/providers/gcp_vertex_gemini.rs | Updated test provider model names and shorthand URLs to 2.5 versions |
| tensorzero-core/tests/e2e/providers/common.rs | Updated model references in test configurations to 2.5 versions |
| tensorzero-core/tests/e2e/db/select_queries.rs | Updated query assertions to match new model names |
| tensorzero-core/tests/e2e/config/tensorzero.models.toml | Updated model definitions and provider configurations to 2.5 versions |
| tensorzero-core/tests/e2e/config/tensorzero.functions.weather_helper.toml | Updated function variant model references to 2.5 |
| tensorzero-core/tests/e2e/config/tensorzero.functions.toml | Updated evaluator model references to 2.5-flash |
| tensorzero-core/tests/e2e/config/tensorzero.functions.json_success.toml | Updated all variant model references to 2.5 versions |
| tensorzero-core/tests/e2e/config/tensorzero.functions.dynamic_json.toml | Updated variant model references to 2.5 versions |
| tensorzero-core/tests/e2e/config/tensorzero.functions.basic_test.toml | Updated variant model references and shorthand URLs to 2.5 |
| tensorzero-core/tests/e2e/best_of_n.rs | Updated test assertions and model name checks to 2.5-flash |
| tensorzero-core/src/providers/gcp_vertex_gemini/optimization.rs | Updated test mock data to reference 2.5-flash-tuned models |
| tensorzero-core/src/providers/gcp_vertex_gemini/mod.rs | Updated unit tests with new model names and shorthand URL examples |
| tensorzero-core/src/config/tests.rs | Updated model configuration test to use 2.5-flash |
| recipes/supervised_fine_tuning/gcp-vertex-gemini/gcp_vertex_gemini_nb.py | Updated default model name to 2.5-flash-lite |
| recipes/supervised_fine_tuning/gcp-vertex-gemini/gcp_vertex_gemini.ipynb | Updated notebook default model name to 2.5-flash-lite |
| examples/guides/providers/google-ai-studio-gemini/config/tensorzero.toml | Updated example configuration to use 2.5-flash-lite |
| examples/guides/providers/gcp-vertex-ai-gemini/config/tensorzero.toml | Updated example configuration to use 2.5-flash |
| docs/integrations/model-providers/google-ai-studio-gemini.mdx | Updated documentation examples to reference 2.5-flash-lite |
| docs/integrations/model-providers/gcp-vertex-ai-gemini.mdx | Updated documentation examples to reference 2.5-flash |
| docs/gateway/configure-models-and-providers.mdx | Updated shorthand example from 2.0-flash-exp to 2.5-flash |
| docs/gateway/configuration-reference.mdx | Updated configuration examples to use 2.5-flash |
| .github/workflows/batch-completion-cron.yml | Updated batch inference workflow to use 2.5-flash |
| supports_batch_inference: false, | ||
| variant_name: "gcp-vertex-gemini-flash-lite-tuned".to_string(), | ||
| model_name: "gemini-2.0-flash-lite-tuned".into(), | ||
| model_name: "gemini-2.5-flash-lite-tuned".into(), |
There was a problem hiding this comment.
The model name was updated to "gemini-2.5-flash-lite-tuned" here, but the corresponding model definition in tensorzero.models.toml still uses the old name "gemini-2.0-flash-lite-tuned". This creates an inconsistency where tests will reference a model name that doesn't exist in the configuration. Either revert this change to keep using "gemini-2.0-flash-lite-tuned" (if the tuned model is intentionally based on 2.0), or update the model definition and all function variant references to use "gemini-2.5-flash-lite-tuned".
Suggested change
| model_name: "gemini-2.5-flash-lite-tuned".into(), | |
| model_name: "gemini-2.0-flash-lite-tuned".into(), |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
No description provided.