Skip to content

feat: route MiniMax through the Anthropic-compatible endpoint and add test coverage #107

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.

Sign up for GitHub

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

Open
octo-patch wants to merge 2 commits into
base: main
Choose a base branch
from
Open

feat: route MiniMax through the Anthropic-compatible endpoint and add test coverage #107

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7a23e49
feat: route MiniMax through the Anthropic-compatible endpoint and add…
octo-patch Aug 4, 2026
18c7707
fix: complete MiniMax routing defaults
octo-patch Aug 11, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
17 changes: 16 additions & 1 deletion .env.example
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,14 +2,24 @@
ANTHROPIC_API_KEY="your-anthropic-api-key" # Needed if proxying *to* Anthropic
OPENAI_API_KEY="sk-..."
GEMINI_API_KEY="your-google-ai-studio-key"
MINIMAX_API_KEY="your-minimax-api-key"

# Optional: Provider Preference and Model Mapping
# Controls which provider (google, openai, or anthropic) is preferred for mapping haiku/sonnet.
# Controls which provider (google, openai, anthropic, or minimax) is preferred for mapping haiku/sonnet.
# Defaults to openai if not set.
# Set to "anthropic" for "just an Anthropic proxy" mode (no remapping)
PREFERRED_PROVIDER="openai"
OPENAI_BASE_URL="https://api.openai.com/v1"

# Optional: MiniMax base URLs.
# OpenAI-compatible endpoint (global); set to https://api.minimaxi.com/v1 for the China region.
MINIMAX_BASE_URL="https://api.minimax.io/v1"
# Anthropic-compatible endpoint (global); set to https://api.minimaxi.com/anthropic for the China region.
MINIMAX_ANTHROPIC_BASE_URL="https://api.minimax.io/anthropic"
# Route MiniMax chat completions through the Anthropic-compatible endpoint by
# default; set to "openai" to use the OpenAI-compatible endpoint instead.
MINIMAX_API_STYLE="anthropic"

# Optional: Specify the exact models to map haiku/sonnet to.
# If PREFERRED_PROVIDER=google, these MUST be valid Gemini model names known to the server.
# Defaults to gemini-2.5-pro and gemini-2.5-flash if PREFERRED_PROVIDER=google.
Expand All @@ -23,6 +33,11 @@ OPENAI_BASE_URL="https://api.openai.com/v1"
# BIG_MODEL="gemini-2.5-pro"
# SMALL_MODEL="gemini-2.5-flash"

# Example MiniMax mapping:
# PREFERRED_PROVIDER="minimax"
# BIG_MODEL="MiniMax-M3" # Optional, it's the default for MiniMax pref
# SMALL_MODEL="MiniMax-M2.7" # Optional, it's the default for MiniMax pref

# Example Google with vertex AI auth via ADC:
# PREFERRED_PROVIDER="google"
# USE_VERTEX_AUTH=true
Expand Down
42 changes: 33 additions & 9 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -42,17 +42,22 @@ A proxy server that lets you use Anthropic clients with Gemini, OpenAI, or Anthr
* `ANTHROPIC_API_KEY`: (Optional) Needed only if proxying *to* Anthropic models.
* `OPENAI_API_KEY`: Your OpenAI API key (Required if using the default OpenAI preference or as fallback).
* `GEMINI_API_KEY`: Your Google AI Studio (Gemini) API key (Required if `PREFERRED_PROVIDER=google` and `USE_VERTEX_AUTH=true`).
* `MINIMAX_API_KEY`: Your MiniMax API key (Required if `PREFERRED_PROVIDER=minimax`).
* `USE_VERTEX_AUTH` (Optional): Set to `true` to use Application Default Credentials (ADC) will be used (no static API key required). Note: when USE_VERTEX_AUTH=true, you must configure `VERTEX_PROJECT` and `VERTEX_LOCATION`.
* `VERTEX_PROJECT` (Optional): Your Google Cloud Project ID (Required if `PREFERRED_PROVIDER=google` and `USE_VERTEX_AUTH=true`).
* `VERTEX_LOCATION` (Optional): The Google Cloud region for Vertex AI (e.g., `us-central1`) (Required if `PREFERRED_PROVIDER=google` and `USE_VERTEX_AUTH=true`).
* `PREFERRED_PROVIDER` (Optional): Set to `openai` (default), `google`, or `anthropic`. This determines the primary backend for mapping `haiku`/`sonnet`.
* `BIG_MODEL` (Optional): The model to map `sonnet` requests to. Defaults to `gpt-4.1` (if `PREFERRED_PROVIDER=openai`) or `gemini-2.5-pro-preview-03-25`. Ignored when `PREFERRED_PROVIDER=anthropic`.
* `SMALL_MODEL` (Optional): The model to map `haiku` requests to. Defaults to `gpt-4.1-mini` (if `PREFERRED_PROVIDER=openai`) or `gemini-2.0-flash`. Ignored when `PREFERRED_PROVIDER=anthropic`.
* `PREFERRED_PROVIDER` (Optional): Set to `openai` (default), `google`, `anthropic`, or `minimax`. This determines the primary backend for mapping `haiku`/`sonnet`.
* `BIG_MODEL` (Optional): The model to map `sonnet` requests to. Defaults to `gpt-4.1`, or `MiniMax-M3` when `PREFERRED_PROVIDER=minimax`. Ignored when `PREFERRED_PROVIDER=anthropic`.
* `SMALL_MODEL` (Optional): The model to map `haiku` requests to. Defaults to `gpt-4.1-mini`, or `MiniMax-M2.7` when `PREFERRED_PROVIDER=minimax`. Ignored when `PREFERRED_PROVIDER=anthropic`.
* `MINIMAX_BASE_URL` (Optional): The MiniMax OpenAI-compatible endpoint. Defaults to `https://api.minimax.io/v1` (global); set to `https://api.minimaxi.com/v1` for the China region.
* `MINIMAX_ANTHROPIC_BASE_URL` (Optional): The MiniMax Anthropic-compatible endpoint. Defaults to `https://api.minimax.io/anthropic` (global); set to `https://api.minimaxi.com/anthropic` for the China region.
* `MINIMAX_API_STYLE` (Optional): Which MiniMax endpoint to route chat completions through. Defaults to `anthropic` (the Anthropic-compatible endpoint); set to `openai` to use the OpenAI-compatible endpoint.

**Mapping Logic:**
- If `PREFERRED_PROVIDER=openai` (default), `haiku`/`sonnet` map to `SMALL_MODEL`/`BIG_MODEL` prefixed with `openai/`.
- If `PREFERRED_PROVIDER=google`, `haiku`/`sonnet` map to `SMALL_MODEL`/`BIG_MODEL` prefixed with `gemini/` *if* those models are in the server's known `GEMINI_MODELS` list (otherwise falls back to OpenAI mapping).
- If `PREFERRED_PROVIDER=anthropic`, `haiku`/`sonnet` requests are passed directly to Anthropic with the `anthropic/` prefix without remapping to different models.
- If `PREFERRED_PROVIDER=minimax`, `haiku`/`sonnet` map to `SMALL_MODEL`/`BIG_MODEL` prefixed with `minimax/`, and bare MiniMax models in the server's known `MINIMAX_MODELS` list are prefixed with `minimax/`. MiniMax requests use the MiniMax API key and are routed through the configured endpoint (the Anthropic-compatible one by default).

4. **Run the server**:
```bash
Expand Down Expand Up @@ -101,12 +106,12 @@ docker run -d --env-file .env -p 8082:8082 ghcr.io/1rgs/claude-code-proxy:latest

## Model Mapping 🗺️

The proxy automatically maps Claude models to either OpenAI or Gemini models based on the configured model:
The proxy automatically maps Claude models to the configured provider:

| Claude Model | Default Mapping | When BIG_MODEL/SMALL_MODEL is a Gemini model |
|--------------|--------------|---------------------------|
| haiku | openai/gpt-4o-mini | gemini/[model-name] |
| sonnet | openai/gpt-4o | gemini/[model-name] |
| Claude Model | OpenAI default | Google mapping | MiniMax default |
|--------------|----------------|----------------|-----------------|
| haiku | openai/gpt-4.1-mini | gemini/[model-name] | minimax/MiniMax-M2.7 |
| sonnet | openai/gpt-4.1 | gemini/[model-name] | minimax/MiniMax-M3 |

### Supported Models

Expand All @@ -130,11 +135,17 @@ The following Gemini models are supported with automatic `gemini/` prefix handli
- gemini-2.5-pro
- gemini-2.5-flash

#### MiniMax Models
The following MiniMax models are supported with automatic `minimax/` prefix handling:
- MiniMax-M3
- MiniMax-M2.7

### Model Prefix Handling
The proxy automatically adds the appropriate prefix to model names:
- OpenAI models get the `openai/` prefix
- Gemini models get the `gemini/` prefix
- The BIG_MODEL and SMALL_MODEL will get the appropriate prefix based on whether they're in the OpenAI or Gemini model lists
- MiniMax models get the `minimax/` prefix
- The BIG_MODEL and SMALL_MODEL use the prefix for the configured provider

For example:
- `gpt-4o` becomes `openai/gpt-4o`
Expand Down Expand Up @@ -185,6 +196,19 @@ PREFERRED_PROVIDER="anthropic"

*Use case: This mode enables you to use the proxy infrastructure (for logging, middleware, request/response processing, etc.) while still using actual Anthropic models rather than being forced to remap to OpenAI or Gemini.*

**Example 3b: Prefer MiniMax**
```dotenv
MINIMAX_API_KEY="your-minimax-key"
PREFERRED_PROVIDER="minimax"
# BIG_MODEL="MiniMax-M3" # Optional, it's the default for MiniMax pref
# SMALL_MODEL="MiniMax-M2.7" # Optional, it's the default for MiniMax pref
# MINIMAX_ANTHROPIC_BASE_URL="https://api.minimax.io/anthropic" # Optional, it's the default (global)
# MINIMAX_BASE_URL="https://api.minimax.io/v1" # Optional, the OpenAI-compatible endpoint
# MINIMAX_API_STYLE="anthropic" # Optional, it's the default; set to "openai" for the OpenAI-compatible endpoint
```

*Use case: MiniMax requests are routed through the MiniMax Anthropic-compatible endpoint by default, so Anthropic clients can talk to MiniMax models without any format translation.*

**Example 4: Use Specific OpenAI Models**
```dotenv
OPENAI_API_KEY="your-openai-key"
Expand Down
Loading