Add bedrock model providers #114
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
…onent for DSE-47182
…flow - Split synthesis_service.py into: * synthesis_service.py (freeform only) * synthesis_legacy_service.py (SFT & Custom_Workflow) - Split evaluator_service.py into: * evaluator_service.py (freeform only) * evaluator_legacy_service.py (SFT & Custom_Workflow) - Updated main.py to route requests to appropriate services - Updated all dependent files (synthesis_job, model_alignment, run scripts) - Created comprehensive test coverage for both legacy and freeform services - Added .gitignore patterns to prevent committing generated data files - Maintained full backward compatibility - all endpoints work as before This refactoring isolates freeform functionality while preserving existing SFT and Custom_Workflow features without breaking changes.
…ervices - Added max_concurrent_topics field to SynthesisRequest (1-100, default: 5) - Added max_workers field to EvaluationRequest (1-100, default: 4) - Updated all synthesis services to use request.max_concurrent_topics - Updated all evaluator services to use request.max_workers - Added validation constraints to prevent invalid values - Updated example payloads in main.py to include new parameters - Services now respect API-configurable concurrency limits while maintaining defaults This allows users to optimize performance based on their infrastructure and workload requirements via API parameters.
- Add CustomEndpointManager for CRUD operations on custom endpoints - Support for all providers: CAII, Bedrock, OpenAI, OpenAI Compatible, Gemini - Complete REST API for endpoint management (POST/GET/PUT/DELETE) - Integration with existing model handlers for automatic credential lookup - Move custom_endpoint_manager to app/core/ for better organization - Add OpenAI_Endpoint_Compatible_Key to environment variables - Fix ImportError: replace Example_eval with EvaluationExample - Restore missing fields: max_concurrent_topics, max_workers, Example_eval, etc. - Add comprehensive API documentation and curl examples
- Remove unnecessary endpoint_id field (use model_id + provider_type as natural unique key) - Remove display_name, created_at, updated_at metadata fields (not needed for server operations) - Remove endpoint_url from Bedrock/OpenAI/Gemini (they use standard APIs) - Keep endpoint_url only for CAII and OpenAI Compatible (custom deployments) - Update API endpoints to use /model_id/provider_type pattern - Simplify JSON storage structure to use model_id:provider_type keys - Test confirmed: server starts and endpoints work correctly
- Implement custom credential lookup for Bedrock provider - Use custom AWS credentials from JSON file when available - Fallback to default AWS credential chain when no custom config found - Update client creation on retry/connection errors - Now all providers (OpenAI, Gemini, CAII, OpenAI Compatible, Bedrock) support custom credentials - Maintain consistent credential priority: JSON > Environment Variables > Error
Add Settings Page Add Models Table Add Model Provider Modal Add Delete Model Add Edit Modal Change Configure Model ID Refetch After Delete
DSE-47416 - Add Model Provider UI
vkeivan
force-pushed
the
add-bedrock-model-providers
branch
5 times, most recently
from
fa7f8da to
bfc1a45
Compare
Add AWS Bedrock Model Provider Fix edit custom model issue
vkeivan
force-pushed
the
add-bedrock-model-providers
branch
from
bfc1a45 to
62421d0
Compare
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.