feat(bailian): add native support for Alibaba Cloud Bailian (百炼)

.env.template

@@ -352,6 +352,11 @@
 # Set base URL to enable (default: http://localhost:11434/v1)
 # OLLAMA_BASE_URL=http://localhost:11434/v1
 
+# Alibaba Cloud Bailian (百炼)
+# BAILIAN_API_KEY=...
+# BAILIAN_BASE_URL=https://dashscope.aliyuncs.com/compatible-mode/v1
+# BAILIAN_MODELS=text-embedding-v3,text-embedding-v4
+
 # vLLM (OpenAI-compatible server)
 # VLLM_API_KEY is optional; set it only if vllm serve was started with --api-key.
 # VLLM_API_KEY=token-abc123

.gitignore

@@ -26,8 +26,14 @@
 # Local git worktrees
 /.worktrees/
 
+# Local pre-commit hook (not shared)
+/.githooks/
+
 # Others
 /*.bck.yml
 /repomix-output.*
 /coverage.out
 /.claude/
+
+# Superpower design docs and plans (never commit)
+/docs/superpowers/

CLAUDE.md

@@ -128,5 +128,5 @@ Full reference: `.env.template` and `config/config.yaml`
 - **Resilience:** Configured via `config/config.yaml` - global `resilience.retry.*` and `resilience.circuit_breaker.*` defaults with optional per-provider overrides under `providers.<name>.resilience.retry.*` and `providers.<name>.resilience.circuit_breaker.*`. Retry defaults: `max_retries` (3), `initial_backoff` (1s), `max_backoff` (30s), `backoff_factor` (2.0), `jitter_factor` (0.1). Circuit breaker defaults: `failure_threshold` (5), `success_threshold` (2), `timeout` (30s)
 - **Metrics:** `METRICS_ENABLED` (false), `METRICS_ENDPOINT` (/metrics)
 - **Guardrails:** Configured via `config/config.yaml` only (except `GUARDRAILS_ENABLED` env var)
-- **Providers:** `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, `GEMINI_API_KEY`, `USE_GOOGLE_GEMINI_NATIVE_API` (true by default; false uses Gemini's OpenAI-compatible chat API), `XAI_API_KEY`, `GROQ_API_KEY`, `OPENROUTER_API_KEY`, `ZAI_API_KEY`, `ZAI_BASE_URL` (optional Z.ai endpoint override), `MINIMAX_API_KEY`, `MINIMAX_BASE_URL` (optional MiniMax endpoint override), `XIAOMI_API_KEY`, `XIAOMI_BASE_URL` (optional Xiaomi MiMo endpoint override), `AZURE_API_KEY`, `AZURE_BASE_URL` (Azure OpenAI deployment base URL), `AZURE_API_VERSION` (optional Azure API version), `ORACLE_API_KEY` (Oracle API key), `ORACLE_BASE_URL` (Oracle OpenAI-compatible base URL), `<PROVIDER>[_SUFFIX]_MODELS` (comma-separated configured model list for any provider type), `OLLAMA_BASE_URL`, `VLLM_BASE_URL`, `VLLM_API_KEY` (optional upstream vLLM bearer token)
+- **Providers:** `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, `GEMINI_API_KEY`, `USE_GOOGLE_GEMINI_NATIVE_API` (true by default; false uses Gemini's OpenAI-compatible chat API), `XAI_API_KEY`, `GROQ_API_KEY`, `OPENROUTER_API_KEY`, `ZAI_API_KEY`, `ZAI_BASE_URL` (optional Z.ai endpoint override), `MINIMAX_API_KEY`, `MINIMAX_BASE_URL` (optional MiniMax endpoint override), `XIAOMI_API_KEY`, `XIAOMI_BASE_URL` (optional Xiaomi MiMo endpoint override), `BAILIAN_API_KEY`, `BAILIAN_BASE_URL` (optional Bailian base URL for region switching; default `https://dashscope.aliyuncs.com/compatible-mode/v1`), `AZURE_API_KEY`, `AZURE_BASE_URL` (Azure OpenAI deployment base URL), `AZURE_API_VERSION` (optional Azure API version), `ORACLE_API_KEY` (Oracle API key), `ORACLE_BASE_URL` (Oracle OpenAI-compatible base URL), `<PROVIDER>[_SUFFIX]_MODELS` (comma-separated configured model list for any provider type), `OLLAMA_BASE_URL`, `VLLM_BASE_URL`, `VLLM_API_KEY` (optional upstream vLLM bearer token)
 - **Provider model metadata:** `providers.<name>.models` accepts either model IDs (strings) or `{id, metadata}` objects. When `metadata` is supplied (`display_name`, `context_window`, `max_output_tokens`, `modes`, `capabilities`, `pricing`, …) it is merged onto the remote ai-model-list entry during enrichment, with operator values winning per-field. Primary use case: advertising context windows, capabilities, and pricing for local models (Ollama) and other custom endpoints whose IDs are not in the upstream registry.

README.md

@@ -100,6 +100,7 @@ Example model identifiers are illustrative and subject to change; consult provid
 | OpenRouter    | `OPENROUTER_API_KEY`                                              | `google/gemini-2.5-flash`          |  ✅  |      ✅      |  ✅   |  ✅   |   ✅    |    ✅    |
 | Z.ai          | `ZAI_API_KEY` (`ZAI_BASE_URL` optional)                           | `glm-5.1`                          |  ✅  |      ✅      |  ✅   |  ❌   |   ❌    |    ✅    |
 | xAI (Grok)    | `XAI_API_KEY`                                                     | `grok-4`                           |  ✅  |      ✅      |  ✅   |  ✅   |   ✅    |    ❌    |
+| Alibaba Cloud Bailian | `BAILIAN_API_KEY` (`BAILIAN_BASE_URL` optional)       | `qwen3-max`                        |  ✅  |      ✅      |  ✅   |  ✅   |   ✅    |    ✅    |
 | MiniMax       | `MINIMAX_API_KEY` (`MINIMAX_BASE_URL` optional)                   | `MiniMax-M3`                       |  ✅  |      ✅      |  ✅   |  ❌   |   ❌    |    ✅    |
 | Xiaomi MiMo   | `XIAOMI_API_KEY` (`XIAOMI_BASE_URL` optional)                     | `mimo-v2.5-pro`                    |  ✅  |      ✅      |  ❌   |  ❌   |   ❌    |    ✅    |
 | Azure OpenAI  | `AZURE_API_KEY` + `AZURE_BASE_URL` (`AZURE_API_VERSION` optional) | `gpt-5`                            |  ✅  |      ✅      |  ✅   |  ✅   |   ✅    |    ✅    |

cmd/gomodel/main.go

@@ -18,6 +18,7 @@ import (
 	"gomodel/internal/providers"
 	"gomodel/internal/providers/anthropic"
 	"gomodel/internal/providers/azure"
+	"gomodel/internal/providers/bailian"
 	"gomodel/internal/providers/bedrock"
 	"gomodel/internal/providers/deepseek"
 	"gomodel/internal/providers/gemini"
@@ -76,7 +77,7 @@ func startApplication(application lifecycleApp, addr string) error {
 
 // @title          GoModel API
 // @version        1.0
-// @description    AI gateway routing requests to multiple LLM providers (OpenAI, Anthropic, Gemini, Groq, OpenRouter, DeepSeek, Z.ai, xAI, MiniMax, Xiaomi MiMo, Oracle, Ollama). Drop-in OpenAI-compatible API.
+// @description    AI gateway routing requests to multiple LLM providers (OpenAI, Anthropic, Gemini, Groq, OpenRouter, DeepSeek, Z.ai, xAI, MiniMax, Xiaomi MiMo, Oracle, Ollama, Bailian). Drop-in OpenAI-compatible API.
 // @BasePath       /
 // @schemes        http
 // @securityDefinitions.apikey BearerAuth
@@ -120,6 +121,7 @@ func main() {
 	factory.Add(openai.Registration)
 	factory.Add(openrouter.Registration)
 	factory.Add(azure.Registration)
+	factory.Add(bailian.Registration)
 	factory.Add(oracle.Registration)
 	factory.Add(anthropic.Registration)
 	factory.Add(bedrock.Registration)

config/config.example.yaml

@@ -13,7 +13,7 @@ server:
   enable_passthrough_routes: true # expose /p/{provider}/{endpoint} passthrough routes
   allow_passthrough_v1_alias: true # allow /p/{provider}/v1/... while keeping /p/{provider}/... canonical
   user_path_header: "X-GoModel-User-Path" # env: USER_PATH_HEADER; inbound header used for user_path scoping
-  enabled_passthrough_providers: ["openai", "anthropic", "openrouter", "zai", "vllm", "deepseek"] # providers enabled on /p/{provider}/...
+  enabled_passthrough_providers: ["openai", "anthropic", "openrouter", "zai", "vllm", "deepseek", "bailian"] # providers enabled on /p/{provider}/...
 
 models:
   enabled_by_default: true # env: MODELS_ENABLED_BY_DEFAULT; when false, models stay unavailable until an override allows one or more user paths
@@ -199,6 +199,15 @@ providers:
     type: anthropic
     api_key: "sk-ant-..."
 
+  bailian:
+    type: bailian
+    api_key: "${BAILIAN_API_KEY}"
+    # base_url: "https://dashscope.aliyuncs.com/compatible-mode/v1"
+    # Alternative regions (replace {workspace-id} with your workspace):
+    # Singapore: "https://{workspace-id}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1"
+    # Frankfurt: "https://{workspace-id}.eu-central-1.maas.aliyuncs.com/compatible-mode/v1"
+    # Hong Kong: "https://{workspace-id}.cn-hongkong.maas.aliyuncs.com/compatible-mode/v1"
+
   gemini:
     type: gemini
     api_key: "..."

config/config_test.go

@@ -1125,7 +1125,7 @@ func TestLoad_ConfigExample_UsesNestedModelCacheSettings(t *testing.T) {
 			t.Fatalf("expected Cache.Model.Redis to be nil in example config, got %+v", result.Config.Cache.Model.Redis)
 		}
 		gotProviders := result.Config.Server.EnabledPassthroughProviders
-		wantProviders := []string{"openai", "anthropic", "openrouter", "zai", "vllm", "deepseek"}
+		wantProviders := []string{"openai", "anthropic", "openrouter", "zai", "vllm", "deepseek", "bailian"}
 		if !reflect.DeepEqual(gotProviders, wantProviders) {
 			t.Fatalf("Server.EnabledPassthroughProviders = %v, want %v", gotProviders, wantProviders)
 		}

docs/docs.json

@@ -104,6 +104,7 @@
                     "providers/anthropic",
                     "providers/gemini",
                     "providers/deepseek",
+"providers/bailian",
                     "providers/xiaomi",
                     "providers/vllm",
                     "providers/multiple-ollama",

docs/openapi.json

@@ -1,7 +1,7 @@
 {
   "openapi": "3.0.0",
   "info": {
-    "description": "AI gateway routing requests to multiple LLM providers (OpenAI, Anthropic, Gemini, Groq, OpenRouter, DeepSeek, Z.ai, xAI, MiniMax, Oracle, Ollama). Drop-in OpenAI-compatible API.",
+    "description": "AI gateway routing requests to multiple LLM providers (OpenAI, Anthropic, Gemini, Groq, OpenRouter, DeepSeek, Z.ai, xAI, MiniMax, Xiaomi MiMo, Bailian, Oracle, Ollama). Drop-in OpenAI-compatible API.",
     "title": "GoModel API",
     "contact": {},
     "version": "1.0"

docs/providers/bailian.mdx

@@ -0,0 +1,104 @@
+---
+title: "Alibaba Cloud Bailian"
+description: "Configure Alibaba Cloud Bailian (百炼 / DashScope) in GoModel, including the max_tokens compatibility shim for models like Qwen."
+icon: "cloud"
+---
+
+Bailian (百炼) is Alibaba Cloud's model-as-a-service platform for the Qwen
+family of models. GoModel routes to Bailian through its OpenAI-compatible
+endpoint (`/compatible-mode/v1`).
+
+Because Bailian deprecated `max_tokens` in April 2026 in favor of
+`max_completion_tokens`, GoModel automatically maps the standard
+`max_tokens` field to `max_completion_tokens` for every request — no
+client change required.
+
+## Configure
+
+```bash
+BAILIAN_API_KEY=...
+```
+
+Or in `config.yaml`:
+
+```yaml
+providers:
+  bailian:
+    type: bailian
+    api_key: "${BAILIAN_API_KEY}"
+    # base_url: "https://dashscope.aliyuncs.com/compatible-mode/v1"
+```
+
+## Base URLs
+
+Bailian's OpenAI-compatible API is available in multiple regions. Set
+`BAILIAN_BASE_URL` to switch:
+
+| Region | URL |
+| ------ | --- |
+| Beijing (default) | `https://dashscope.aliyuncs.com/compatible-mode/v1` |
+| Singapore | `https://{workspace-id}.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1` |
+| Frankfurt | `https://{workspace-id}.eu-central-1.maas.aliyuncs.com/compatible-mode/v1` |
+| Hong Kong | `https://{workspace-id}.cn-hongkong.maas.aliyuncs.com/compatible-mode/v1` |
+
+## Model IDs
+
+Common Qwen model identifiers — check the [Bailian model
+list](https://www.alibabacloud.com/help/zh/model-studio/model-list) for the
+current catalog:
+
+| Model | Example ID |
+| ----- | ---------- |
+| Qwen 3.7 Max | `qwen3.7-max` |
+| Qwen 3.7 Plus | `qwen3.7-plus` |
+| Qwen 3.6 Flash | `qwen3.6-flash` |
+| Qwen 3 Max | `qwen3-max` |
+| Qwen 3 Plus | `qwen3-plus` |
+| Qwen 3 Flash | `qwen3-flash` |
+| Qwen 3 Coder Plus | `qwen3-coder-plus` |
+| Text Embedding | `text-embedding-v3` |
+
+## `max_tokens` compatibility
+
+Bailian deprecated `max_tokens` on 2026-04-20 (effective 2026-05-30).
+All Bailian models now require `max_completion_tokens` instead.
+
+GoModel transparently maps the standard `max_tokens` parameter to
+`max_completion_tokens` for every bailian model — send `max_tokens` as you
+normally would, and GoModel rewrites it before forwarding to Bailian.
+
+```bash
+# max_tokens=4096 is automatically sent as max_completion_tokens=4096
+curl http://localhost:8080/v1/chat/completions \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "qwen3-max",
+    "messages": [{"role": "user", "content": "Hello"}],
+    "max_tokens": 4096
+  }'
+```
+
+## Supported features
+
+| Feature | Supported |
+| ------- | :-------: |
+| Chat completions | ✅ |
+| Streaming chat | ✅ |
+| Responses (`/v1/responses`) | ✅ (translated to chat) |
+| Embeddings | ✅ (configure model IDs via `BAILIAN_MODELS`) |
+| Files (`/v1/files`) | ✅ |
+| Batches (`/v1/batches`) | ✅ |
+| Passthrough (`/p/bailian/...`) | ✅ |
+
+<Note>
+  Embedding models (`text-embedding-v3`, `text-embedding-v4`) are served by
+  the compatible-mode API but are **not** auto-discovered from the upstream
+  `/v1/models` endpoint. Set `BAILIAN_MODELS=text-embedding-v3,text-embedding-v4`
+  or use `CONFIGURED_PROVIDER_MODELS_MODE=allowlist` to make them available.
+</Note>
+
+## References
+
+- [Bailian documentation](https://www.alibabacloud.com/help/zh/model-studio/)
+- [OpenAI-compatible API reference](https://www.alibabacloud.com/help/zh/model-studio/compatibility-with-openai-responses-api)
+- [Qwen model list](https://www.alibabacloud.com/help/zh/model-studio/model-list)

docs/providers/overview.mdx

@@ -25,6 +25,7 @@ quirks.
 | Z.ai | `ZAI_API_KEY` (`ZAI_BASE_URL` optional) | — |
 | xAI (Grok) | `XAI_API_KEY` | — |
 | MiniMax | `MINIMAX_API_KEY` (`MINIMAX_BASE_URL` optional) | — |
+| Alibaba Cloud Bailian | `BAILIAN_API_KEY` (`BAILIAN_BASE_URL` optional) | [Alibaba Cloud Bailian](/providers/bailian) |
 | Xiaomi MiMo | `XIAOMI_API_KEY` (`XIAOMI_BASE_URL` optional) | [Xiaomi MiMo](/providers/xiaomi) |
 | Azure OpenAI | `AZURE_API_KEY` + `AZURE_BASE_URL` (`AZURE_API_VERSION` optional) | [Azure OpenAI](/providers/azure) |
 | Amazon Bedrock | `BEDROCK_BASE_URL` (region or endpoint) + AWS credentials | [Amazon Bedrock](/providers/bedrock) |