feat: add MiniMax as LLM provider

Add MiniMax (M2.7, M2.5, M2.5-highspeed) as a first-class LLM provider
using the OpenAI-compatible API at api.minimax.io/v1.

- New MiniMaxEngine extending OpenAiEngine with think-tag stripping
- Full setup wizard integration (provider selection, model list, API key URL)
- Dynamic model fetching via /v1/models endpoint with 7-day cache
- Error handling with billing URL and model suggestions
- README updated with MiniMax provider config example
- 32 unit tests + 3 integration tests

Author: PR Bot

README.md

@@ -122,7 +122,7 @@ Create a `.env` file and add OpenCommit config variables there like this:
 
 ```env
 ...
-OCO_AI_PROVIDER=<openai (default), anthropic, azure, ollama, gemini, flowise, deepseek, aimlapi>
+OCO_AI_PROVIDER=<openai (default), anthropic, azure, ollama, gemini, flowise, deepseek, aimlapi, minimax>
 OCO_API_KEY=<your OpenAI API token> // or other LLM provider API token
 OCO_API_URL=<may be used to set proxy path to OpenAI api>
 OCO_API_CUSTOM_HEADERS=<JSON string of custom HTTP headers to include in API requests>
@@ -235,6 +235,8 @@ oco config set OCO_AI_PROVIDER=azure OCO_API_KEY=<your_azure_api_key> OCO_API_UR
 oco config set OCO_AI_PROVIDER=flowise OCO_API_KEY=<your_flowise_api_key> OCO_API_URL=<your_flowise_endpoint>
 
 oco config set OCO_AI_PROVIDER=ollama OCO_API_KEY=<your_ollama_api_key> OCO_API_URL=<your_ollama_endpoint>
+
+oco config set OCO_AI_PROVIDER=minimax OCO_API_KEY=<your_minimax_api_key>
 ```
 
 ### Locale configuration

src/commands/config.ts

@@ -136,6 +136,8 @@ export const MODEL_LIST = {
   ],
   deepseek: ['deepseek-chat', 'deepseek-reasoner'],
 
+  minimax: ['MiniMax-M2.7', 'MiniMax-M2.5', 'MiniMax-M2.5-highspeed'],
+
   // AI/ML API available chat-completion models
   // https://api.aimlapi.com/v1/models
   aimlapi: [
@@ -593,6 +595,8 @@ const getDefaultModel = (provider: string | undefined): string => {
       return MODEL_LIST.aimlapi[0];
     case 'openrouter':
       return MODEL_LIST.openrouter[0];
+    case 'minimax':
+      return MODEL_LIST.minimax[0];
     default:
       return MODEL_LIST.openai[0];
   }
@@ -784,9 +788,10 @@ export const configValidators = {
         'groq',
         'deepseek',
         'aimlapi',
-        'openrouter'
+        'openrouter',
+        'minimax'
       ].includes(value) || value.startsWith('ollama'),
-      `${value} is not supported yet, use 'ollama', 'mlx', 'anthropic', 'azure', 'gemini', 'flowise', 'mistral', 'deepseek', 'aimlapi' or 'openai' (default)`
+      `${value} is not supported yet, use 'ollama', 'mlx', 'anthropic', 'azure', 'gemini', 'flowise', 'mistral', 'deepseek', 'aimlapi', 'minimax' or 'openai' (default)`
     );
 
     return value;
@@ -844,7 +849,8 @@ export enum OCO_AI_PROVIDER_ENUM {
   MLX = 'mlx',
   DEEPSEEK = 'deepseek',
   AIMLAPI = 'aimlapi',
-  OPENROUTER = 'openrouter'
+  OPENROUTER = 'openrouter',
+  MINIMAX = 'minimax'
 }
 
 export const PROVIDER_API_KEY_URLS: Record<string, string | null> = {
@@ -857,6 +863,7 @@ export const PROVIDER_API_KEY_URLS: Record<string, string | null> = {
   [OCO_AI_PROVIDER_ENUM.OPENROUTER]: 'https://openrouter.ai/keys',
   [OCO_AI_PROVIDER_ENUM.AIMLAPI]: 'https://aimlapi.com/app/keys',
   [OCO_AI_PROVIDER_ENUM.AZURE]: 'https://portal.azure.com/',
+  [OCO_AI_PROVIDER_ENUM.MINIMAX]: 'https://platform.minimaxi.com/user-center/basic-information/interface-key',
   [OCO_AI_PROVIDER_ENUM.OLLAMA]: null,
   [OCO_AI_PROVIDER_ENUM.MLX]: null,
   [OCO_AI_PROVIDER_ENUM.FLOWISE]: null,
@@ -871,7 +878,8 @@ export const RECOMMENDED_MODELS: Record<string, string> = {
   [OCO_AI_PROVIDER_ENUM.MISTRAL]: 'mistral-small-latest',
   [OCO_AI_PROVIDER_ENUM.DEEPSEEK]: 'deepseek-chat',
   [OCO_AI_PROVIDER_ENUM.OPENROUTER]: 'openai/gpt-4o-mini',
-  [OCO_AI_PROVIDER_ENUM.AIMLAPI]: 'gpt-4o-mini'
+  [OCO_AI_PROVIDER_ENUM.AIMLAPI]: 'gpt-4o-mini',
+  [OCO_AI_PROVIDER_ENUM.MINIMAX]: 'MiniMax-M2.7'
 }
 
 export type ConfigType = {

src/commands/setup.ts

@@ -31,7 +31,8 @@ const PROVIDER_DISPLAY_NAMES: Record<string, string> = {
   [OCO_AI_PROVIDER_ENUM.OPENROUTER]: 'OpenRouter (Multiple providers)',
   [OCO_AI_PROVIDER_ENUM.AIMLAPI]: 'AI/ML API',
   [OCO_AI_PROVIDER_ENUM.AZURE]: 'Azure OpenAI',
-  [OCO_AI_PROVIDER_ENUM.MLX]: 'MLX (Apple Silicon, local)'
+  [OCO_AI_PROVIDER_ENUM.MLX]: 'MLX (Apple Silicon, local)',
+  [OCO_AI_PROVIDER_ENUM.MINIMAX]: 'MiniMax (M2.7, M2.5, fast inference)'
 };
 
 const PRIMARY_PROVIDERS = [
@@ -48,7 +49,8 @@ const OTHER_PROVIDERS = [
   OCO_AI_PROVIDER_ENUM.OPENROUTER,
   OCO_AI_PROVIDER_ENUM.AIMLAPI,
   OCO_AI_PROVIDER_ENUM.AZURE,
-  OCO_AI_PROVIDER_ENUM.MLX
+  OCO_AI_PROVIDER_ENUM.MLX,
+  OCO_AI_PROVIDER_ENUM.MINIMAX
 ];
 
 const NO_API_KEY_PROVIDERS = [

src/engine/minimax.ts

@@ -0,0 +1,49 @@
+import { OpenAI } from 'openai';
+import { GenerateCommitMessageErrorEnum } from '../generateCommitMessageFromGitDiff';
+import { normalizeEngineError } from '../utils/engineErrorHandler';
+import { removeContentTags } from '../utils/removeContentTags';
+import { tokenCount } from '../utils/tokenCount';
+import { OpenAiEngine, OpenAiConfig } from './openAi';
+
+export interface MiniMaxConfig extends OpenAiConfig {}
+
+export class MiniMaxEngine extends OpenAiEngine {
+  constructor(config: MiniMaxConfig) {
+    super({
+      baseURL: 'https://api.minimax.io/v1',
+      ...config
+    });
+  }
+
+  public generateCommitMessage = async (
+    messages: Array<OpenAI.Chat.Completions.ChatCompletionMessageParam>
+  ): Promise<string | null> => {
+    const params = {
+      model: this.config.model,
+      messages,
+      temperature: 0.01,
+      top_p: 0.1,
+      max_tokens: this.config.maxTokensOutput
+    };
+
+    try {
+      const REQUEST_TOKENS = messages
+        .map((msg) => tokenCount(msg.content as string) + 4)
+        .reduce((a, b) => a + b, 0);
+
+      if (
+        REQUEST_TOKENS >
+        this.config.maxTokensInput - this.config.maxTokensOutput
+      )
+        throw new Error(GenerateCommitMessageErrorEnum.tooMuchTokens);
+
+      const completion = await this.client.chat.completions.create(params);
+
+      const message = completion.choices[0].message;
+      let content = message?.content;
+      return removeContentTags(content, 'think');
+    } catch (error) {
+      throw normalizeEngineError(error, 'minimax', this.config.model);
+    }
+  };
+}

src/utils/engine.ts

@@ -13,6 +13,7 @@ import { MLXEngine } from '../engine/mlx';
 import { DeepseekEngine } from '../engine/deepseek';
 import { AimlApiEngine } from '../engine/aimlapi';
 import { OpenRouterEngine } from '../engine/openrouter';
+import { MiniMaxEngine } from '../engine/minimax';
 
 export function parseCustomHeaders(headers: any): Record<string, string> {
   let parsedHeaders = {};
@@ -88,6 +89,9 @@ export function getEngine(): AiEngine {
     case OCO_AI_PROVIDER_ENUM.OPENROUTER:
       return new OpenRouterEngine(DEFAULT_CONFIG);
 
+    case OCO_AI_PROVIDER_ENUM.MINIMAX:
+      return new MiniMaxEngine(DEFAULT_CONFIG);
+
     default:
       return new OpenAiEngine(DEFAULT_CONFIG);
   }

src/utils/errors.ts

@@ -12,6 +12,7 @@ export const PROVIDER_BILLING_URLS: Record<string, string | null> = {
   [OCO_AI_PROVIDER_ENUM.OPENROUTER]: 'https://openrouter.ai/credits',
   [OCO_AI_PROVIDER_ENUM.AIMLAPI]: 'https://aimlapi.com/app/billing',
   [OCO_AI_PROVIDER_ENUM.AZURE]: 'https://portal.azure.com/#view/Microsoft_Azure_CostManagement',
+  [OCO_AI_PROVIDER_ENUM.MINIMAX]: 'https://platform.minimaxi.com/user-center/basic-information',
   [OCO_AI_PROVIDER_ENUM.OLLAMA]: null,
   [OCO_AI_PROVIDER_ENUM.MLX]: null,
   [OCO_AI_PROVIDER_ENUM.FLOWISE]: null,
@@ -202,6 +203,8 @@ export function getRecommendedModel(provider: string): string | null {
       return 'openai/gpt-4o-mini';
     case OCO_AI_PROVIDER_ENUM.AIMLAPI:
       return 'gpt-4o-mini';
+    case OCO_AI_PROVIDER_ENUM.MINIMAX:
+      return 'MiniMax-M2.7';
     default:
       return null;
   }

src/utils/modelCache.ts

@@ -185,6 +185,30 @@ export async function fetchOpenRouterModels(apiKey: string): Promise<string[]> {
   }
 }
 
+export async function fetchMiniMaxModels(apiKey: string): Promise<string[]> {
+  try {
+    const response = await fetch('https://api.minimax.io/v1/models', {
+      headers: {
+        Authorization: `Bearer ${apiKey}`
+      }
+    });
+
+    if (!response.ok) {
+      return MODEL_LIST.minimax;
+    }
+
+    const data = await response.json();
+    const models = data.data
+      ?.map((m: { id: string }) => m.id)
+      .filter((id: string) => id.startsWith('MiniMax-'))
+      .sort();
+
+    return models && models.length > 0 ? models : MODEL_LIST.minimax;
+  } catch {
+    return MODEL_LIST.minimax;
+  }
+}
+
 export async function fetchDeepSeekModels(apiKey: string): Promise<string[]> {
   try {
     const response = await fetch('https://api.deepseek.com/v1/models', {
@@ -273,6 +297,14 @@ export async function fetchModelsForProvider(
       }
       break;
 
+    case OCO_AI_PROVIDER_ENUM.MINIMAX:
+      if (apiKey) {
+        models = await fetchMiniMaxModels(apiKey);
+      } else {
+        models = MODEL_LIST.minimax;
+      }
+      break;
+
     case OCO_AI_PROVIDER_ENUM.AIMLAPI:
       models = MODEL_LIST.aimlapi;
       break;

test/unit/minimax-integration.test.ts

@@ -0,0 +1,112 @@
+import { OpenAI } from 'openai';
+
+// Mock @clack/prompts to prevent process.exit calls
+jest.mock('@clack/prompts', () => ({
+  intro: jest.fn(),
+  outro: jest.fn()
+}));
+
+/**
+ * Integration tests for MiniMax engine.
+ * These tests verify the MiniMax API works correctly via OpenAI-compatible SDK.
+ * This mirrors the exact behavior of MiniMaxEngine which extends OpenAiEngine.
+ *
+ * Run with: MINIMAX_API_KEY=<key> npm run test -- test/unit/minimax-integration.test.ts
+ */
+const MINIMAX_API_KEY = process.env.MINIMAX_API_KEY;
+const describeIntegration = MINIMAX_API_KEY ? describe : describe.skip;
+
+describeIntegration('MiniMax Integration (requires MINIMAX_API_KEY)', () => {
+  let client: OpenAI;
+
+  beforeAll(() => {
+    client = new OpenAI({
+      apiKey: MINIMAX_API_KEY!,
+      baseURL: 'https://api.minimax.io/v1'
+    });
+  });
+
+  it('should generate a commit message with M2.7', async () => {
+    const completion = await client.chat.completions.create({
+      model: 'MiniMax-M2.7',
+      messages: [
+        {
+          role: 'system',
+          content:
+            'You are an expert at writing concise, meaningful git commit messages. Generate a conventional commit message for the provided code diff. Output only the commit message, nothing else.'
+        },
+        {
+          role: 'user',
+          content: `diff --git a/src/utils.ts b/src/utils.ts
+--- a/src/utils.ts
++++ b/src/utils.ts
+@@ -10,6 +10,10 @@ export function formatDate(date: Date): string {
+   return date.toISOString();
+ }
+
++export function formatCurrency(amount: number, currency: string = 'USD'): string {
++  return new Intl.NumberFormat('en-US', { style: 'currency', currency }).format(amount);
++}
++
+ export function capitalize(str: string): string {`
+        }
+      ],
+      temperature: 0.01,
+      top_p: 0.1,
+      max_tokens: 500
+    });
+
+    const content = completion.choices[0].message?.content;
+    expect(content).toBeDefined();
+    expect(typeof content).toBe('string');
+    expect(content!.length).toBeGreaterThan(0);
+  }, 30000);
+
+  it('should generate commit message with M2.5-highspeed', async () => {
+    const completion = await client.chat.completions.create({
+      model: 'MiniMax-M2.5-highspeed',
+      messages: [
+        {
+          role: 'system',
+          content:
+            'You are an expert at writing concise git commit messages. Generate a conventional commit message. Output only the commit message.'
+        },
+        {
+          role: 'user',
+          content: `diff --git a/README.md b/README.md
+--- a/README.md
++++ b/README.md
+@@ -1,3 +1,5 @@
+ # My Project
+
+ A simple project.
++
++## Installation`
+        }
+      ],
+      temperature: 0.01,
+      top_p: 0.1,
+      max_tokens: 500
+    });
+
+    const content = completion.choices[0].message?.content;
+    expect(content).toBeDefined();
+    expect(typeof content).toBe('string');
+    expect(content!.length).toBeGreaterThan(0);
+  }, 30000);
+
+  it('should handle authentication error with invalid API key', async () => {
+    const badClient = new OpenAI({
+      apiKey: 'invalid-api-key',
+      baseURL: 'https://api.minimax.io/v1'
+    });
+
+    await expect(
+      badClient.chat.completions.create({
+        model: 'MiniMax-M2.7',
+        messages: [{ role: 'user', content: 'test' }],
+        max_tokens: 10
+      })
+    ).rejects.toThrow();
+  }, 30000);
+});