hobokenchicken
6b10d4249c
This commit replaces the Axum/Rust backend with a Gin/Go implementation. The original Rust code has been archived in the 'rust' branch.
49 lines
1.6 KiB
Markdown
49 lines
1.6 KiB
Markdown
# Migration TODO List
|
|
|
|
## Completed Tasks
|
|
- [x] Initial Go project setup
|
|
- [x] Database schema & migrations
|
|
- [x] Configuration loader (Viper)
|
|
- [x] Auth Middleware
|
|
- [x] Basic Provider implementations (OpenAI, Gemini, DeepSeek, Grok)
|
|
- [x] Streaming Support (SSE & Gemini custom streaming)
|
|
- [x] Move Rust files to `rust_backup`
|
|
- [x] Enhanced `helpers.go` for Multimodal & Tool Calling (OpenAI compatible)
|
|
- [x] Enhanced `server.go` for robust request conversion
|
|
- [x] Dashboard Management APIs (Clients, Tokens, Users, Providers)
|
|
- [x] Dashboard Analytics & Usage Summary
|
|
- [x] WebSocket for real-time dashboard updates
|
|
|
|
## Feature Parity Checklist (High Priority)
|
|
|
|
### OpenAI Provider
|
|
- [x] Tool Calling
|
|
- [x] Multimodal (Images) support
|
|
- [ ] Reasoning Content (CoT) support for `o1`, `o3` (need to ensure it's parsed in responses)
|
|
- [ ] Support for `/v1/responses` API (required for some gpt-5/o1 models)
|
|
|
|
### Gemini Provider
|
|
- [x] Tool Calling (mapping to Gemini format)
|
|
- [x] Multimodal (Images) support
|
|
- [x] Reasoning/Thought support
|
|
- [x] Handle Tool Response role in unified format
|
|
|
|
### DeepSeek Provider
|
|
- [x] Reasoning Content (CoT) support
|
|
- [x] Parameter sanitization for `deepseek-reasoner`
|
|
- [x] Tool Calling support
|
|
|
|
### Grok Provider
|
|
- [x] Tool Calling support
|
|
- [x] Multimodal support
|
|
|
|
## Infrastructure & Middleware
|
|
- [ ] Implement Request Logging to SQLite (asynchronous)
|
|
- [ ] Implement Rate Limiting (`golang.org/x/time/rate`)
|
|
- [ ] Implement Circuit Breaker (`github.com/sony/gobreaker`)
|
|
- [ ] Implement Model Cost Calculation logic
|
|
|
|
## Verification
|
|
- [ ] Unit tests for feature-specific mapping (CoT, Tools, Images)
|
|
- [ ] Integration tests with live LLM APIs
|