feat: 实现模型别名系统，支持跨渠道统一模型名称 #2311
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
- Add comprehensive model alias mapping system in relay/model/alias.go - Support standard model names (gpt-4o, claude-3-sonnet) across different channels - Modify AddAbilities() to create aliases for both standard and channel-specific names - Update text processing to resolve aliases before channel-specific mapping - Enhance billing system to support alias-based model ratio lookup - Add comprehensive tests for alias resolution and reverse mapping - Support major providers: OpenRouter, Anthropic, Gemini, Groq This allows users to use consistent model names (e.g., 'gpt-4o') regardless of channel provider, with automatic mapping to channel-specific names (e.g., 'openai/gpt-4o' for OpenRouter).
There was a problem hiding this comment.
Pull Request Overview
This PR implements a comprehensive model alias system that allows mapping between standard model names and channel-specific model names across multiple AI service providers. The system provides bidirectional mapping capabilities and integrates alias resolution into the request processing pipeline.
Key changes:
- Core alias mapping system with support for OpenRouter, Anthropic, Gemini, and Groq channels
- Integration of alias resolution in request processing (before existing model mapping)
- Enhanced billing system to support alias-based rate lookups
- Automatic ability registration for both standard and channel-specific model names
Reviewed Changes
Copilot reviewed 6 out of 6 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| relay/model/alias.go | Core alias mapping system with thread-safe operations and comprehensive model mappings |
| relay/model/alias_test.go | Complete test suite covering all alias functionality with benchmarks |
| relay/controller/text.go | Integration of alias resolution in text request processing |
| relay/controller/helper.go | Helper functions for lightweight alias resolution to avoid circular imports |
| relay/billing/ratio/model.go | Enhanced billing system to support alias-based model rate lookups |
| model/ability.go | Automatic expansion of model abilities to include standard names for channel-specific models |
Comments suppressed due to low confidence (2)
relay/controller/helper.go:1
- Magic numbers are used for channel types instead of constants from the channeltype package. This makes the code harder to maintain and understand. Consider using channeltype.OpenRouter, channeltype.Anthropic, etc.
package controller
relay/billing/ratio/model.go:1
- Magic numbers are used for channel types instead of constants from the channeltype package. This makes the code harder to maintain and understand. Consider using channeltype.OpenRouter, channeltype.Anthropic, etc.
package ratio
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
| // This is a lightweight version to avoid importing the full alias module | ||
| func getModelAliasesForChannelType(channelType int) map[string]string { | ||
| switch channelType { | ||
| case 24: // OpenRouter |
There was a problem hiding this comment.
Magic numbers are used for channel types instead of constants from the channeltype package. This makes the code harder to maintain and understand. Consider using channeltype.OpenRouter, etc.
Suggested change
| case 24: // OpenRouter | |
| case channeltype.OpenRouter: // OpenRouter |
Comment on lines
+167
to
+210
| func getChannelModelAliases(channelType int) map[string]string { | ||
| switch channelType { | ||
| case 24: // OpenRouter | ||
| return map[string]string{ | ||
| "gpt-4o": "openai/gpt-4o", | ||
| "gpt-4o-mini": "openai/gpt-4o-mini", | ||
| "gpt-4": "openai/gpt-4", | ||
| "gpt-4-turbo": "openai/gpt-4-turbo", | ||
| "gpt-3.5-turbo": "openai/gpt-3.5-turbo", | ||
| "gpt-3.5-turbo-0125": "openai/gpt-3.5-turbo-0125", | ||
| "o1": "openai/o1", | ||
| "o1-mini": "openai/o1-mini", | ||
| "o1-preview": "openai/o1-preview", | ||
| "claude-3-haiku": "anthropic/claude-3-haiku", | ||
| "claude-3-sonnet": "anthropic/claude-3-sonnet", | ||
| "claude-3-opus": "anthropic/claude-3-opus", | ||
| "claude-3.5-sonnet": "anthropic/claude-3.5-sonnet", | ||
| "claude-3.5-haiku": "anthropic/claude-3.5-haiku", | ||
| } | ||
| case 18: // Anthropic | ||
| return map[string]string{ | ||
| "claude-3-haiku": "claude-3-haiku-20240307", | ||
| "claude-3-sonnet": "claude-3-sonnet-20240229", | ||
| "claude-3-opus": "claude-3-opus-20240229", | ||
| "claude-3.5-sonnet": "claude-3-5-sonnet-20241022", | ||
| "claude-3.5-haiku": "claude-3-5-haiku-20241022", | ||
| } | ||
| case 28: // Gemini | ||
| return map[string]string{ | ||
| "gemini-pro": "gemini-pro", | ||
| "gemini-pro-1.5": "gemini-1.5-pro-latest", | ||
| "gemini-flash-1.5": "gemini-1.5-flash-latest", | ||
| } | ||
| case 33: // Groq | ||
| return map[string]string{ | ||
| "llama-3-8b-instruct": "llama3-8b-8192", | ||
| "llama-3-70b-instruct": "llama3-70b-8192", | ||
| "llama-3.1-8b-instruct": "llama-3.1-8b-instant", | ||
| "llama-3.1-70b-instruct": "llama-3.1-70b-versatile", | ||
| } | ||
| default: | ||
| return map[string]string{} | ||
| } | ||
| } |
There was a problem hiding this comment.
This function duplicates the alias mappings that already exist in relay/model/alias.go. This creates a maintenance burden where aliases need to be updated in multiple places. Consider refactoring to use a single source of truth for alias mappings.
Comment on lines
+754
to
+776
| switch channelType { | ||
| case 24: // OpenRouter | ||
| return map[string]string{ | ||
| "gpt-4o": "openai/gpt-4o", | ||
| "gpt-4o-mini": "openai/gpt-4o-mini", | ||
| "gpt-4": "openai/gpt-4", | ||
| "gpt-4-turbo": "openai/gpt-4-turbo", | ||
| "gpt-3.5-turbo": "openai/gpt-3.5-turbo", | ||
| "claude-3-haiku": "anthropic/claude-3-haiku", | ||
| "claude-3-sonnet": "anthropic/claude-3-sonnet", | ||
| "claude-3-opus": "anthropic/claude-3-opus", | ||
| "claude-3.5-sonnet": "anthropic/claude-3.5-sonnet", | ||
| } | ||
| case 18: // Anthropic | ||
| return map[string]string{ | ||
| "claude-3-haiku": "claude-3-haiku-20240307", | ||
| "claude-3-sonnet": "claude-3-sonnet-20240229", | ||
| "claude-3-opus": "claude-3-opus-20240229", | ||
| "claude-3.5-sonnet": "claude-3-5-sonnet-20241022", | ||
| } | ||
| default: | ||
| return map[string]string{} | ||
| } |
There was a problem hiding this comment.
This function duplicates the alias mappings that already exist in relay/model/alias.go. This creates a maintenance burden where aliases need to be updated in multiple places. Consider refactoring to use a single source of truth for alias mappings.
Suggested change
| switch channelType { | |
| case 24: // OpenRouter | |
| return map[string]string{ | |
| "gpt-4o": "openai/gpt-4o", | |
| "gpt-4o-mini": "openai/gpt-4o-mini", | |
| "gpt-4": "openai/gpt-4", | |
| "gpt-4-turbo": "openai/gpt-4-turbo", | |
| "gpt-3.5-turbo": "openai/gpt-3.5-turbo", | |
| "claude-3-haiku": "anthropic/claude-3-haiku", | |
| "claude-3-sonnet": "anthropic/claude-3-sonnet", | |
| "claude-3-opus": "anthropic/claude-3-opus", | |
| "claude-3.5-sonnet": "anthropic/claude-3.5-sonnet", | |
| } | |
| case 18: // Anthropic | |
| return map[string]string{ | |
| "claude-3-haiku": "claude-3-haiku-20240307", | |
| "claude-3-sonnet": "claude-3-sonnet-20240229", | |
| "claude-3-opus": "claude-3-opus-20240229", | |
| "claude-3.5-sonnet": "claude-3-5-sonnet-20241022", | |
| } | |
| default: | |
| return map[string]string{} | |
| } | |
| // Use the alias mapping from relay/model/alias.go as the single source of truth | |
| return model.GetAliasMap(channelType) |
Comment on lines
+121
to
+142
| func getModelAliasesForChannelType(channelType int) map[string]string { | ||
| switch channelType { | ||
| case 24: // OpenRouter | ||
| return map[string]string{ | ||
| "gpt-4o": "openai/gpt-4o", | ||
| "gpt-4o-mini": "openai/gpt-4o-mini", | ||
| "gpt-4": "openai/gpt-4", | ||
| "gpt-4-turbo": "openai/gpt-4-turbo", | ||
| "gpt-3.5-turbo": "openai/gpt-3.5-turbo", | ||
| "o1": "openai/o1", | ||
| "o1-mini": "openai/o1-mini", | ||
| "o1-preview": "openai/o1-preview", | ||
| "claude-3-haiku": "anthropic/claude-3-haiku", | ||
| "claude-3-sonnet": "anthropic/claude-3-sonnet", | ||
| "claude-3-opus": "anthropic/claude-3-opus", | ||
| "claude-3.5-sonnet": "anthropic/claude-3.5-sonnet", | ||
| "claude-3.5-haiku": "anthropic/claude-3.5-haiku", | ||
| } | ||
| default: | ||
| return map[string]string{} | ||
| } | ||
| } |
There was a problem hiding this comment.
This function duplicates the alias mappings that already exist in relay/model/alias.go. This creates a maintenance burden where aliases need to be updated in multiple places. Consider refactoring to use a single source of truth for alias mappings.
iceyarmu
pushed a commit
to iceyarmu/one-api
that referenced
this pull request
Dec 3, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
功能特性:
支持的模型映射示例:
close #2310