feat: Add Apertis provider support

packages/opencode/src/provider/provider.ts

@@ -504,6 +504,17 @@ export namespace Provider {
         },
       }
     },
+    apertis: async () => {
+      return {
+        autoload: false,
+        options: {
+          headers: {
+            "HTTP-Referer": "https://opencode.ai/",
+            "X-Title": "opencode",
+          },
+        },
+      }
+    },
   }
 
   export const Model = z

packages/web/src/content/docs/providers.mdx

@@ -122,6 +122,106 @@ Don't see a provider here? Submit a PR.
 
 ---
 
+### Apertis
+
+Apertis provides access to 12+ state-of-the-art models including Claude Opus 4.5, GPT-5.2, Gemini 3 Pro, and specialized thinking models for enhanced reasoning capabilities.
+
+1. Head over to [Apertis](https://apertis.ai/), create an account, and generate an API key.
+
+2. Run the `/connect` command and search for **Apertis**.
+
+   ```txt
+   /connect
+   ```
+
+3. Enter your Apertis API key.
+
+   ```txt
+   ┌ API key
+   │
+   │
+   └ enter
+   ```
+
+4. Run the `/models` command to select a model.
+
+   ```txt
+   /models
+   ```
+
+   Available models include:
+   - **Claude Opus 4.5** & **Claude Sonnet 4.5** - Latest Anthropic models with extended thinking variants
+   - **GPT-5.2** series - Including Pro, standard, and Codex variants for code generation
+   - **Gemini 3 Pro Preview** - Google's next-generation multimodal model
+   - **GLM-4.7** - Advanced multilingual model with thinking mode
+   - **Grok Code Fast 1** - Speed-optimized code generation
+   - **Minimax M2.1** - High-performance general-purpose model
+
+#### Advanced Configuration
+
+You can customize Apertis models through your opencode config:
+
+```json title="opencode.json"
+{
+  "$schema": "https://opencode.ai/config.json",
+  "provider": {
+    "apertis": {
+      "models": {
+        "claude-opus-4-5-20251101-thinking": {
+          "options": {
+            "maxTokens": 8192,
+            "temperature": 0.7
+          }
+        }
+      }
+    }
+  }
+}
+```
+
+#### Using Thinking Models
+
+Apertis offers specialized thinking models (indicated by `-thinking` suffix) that provide enhanced reasoning capabilities:
+
+```json title="opencode.json"
+{
+  "$schema": "https://opencode.ai/config.json",
+  "provider": {
+    "apertis": {
+      "models": {
+        "claude-sonnet-4.5-20250929-thinking": {},
+        "glm-4.7-thinking": {}
+      }
+    }
+  }
+}
+```
+
+These models excel at:
+- Complex problem-solving and multi-step reasoning
+- Mathematical and logical analysis
+- Code architecture and design decisions
+- Detailed technical explanations
+
+#### Troubleshooting
+
+**Connection Issues:**
+- Verify your API key is valid at [Apertis Dashboard](https://apertis.ai/)
+- Check that the API key has not expired or been revoked
+- Ensure your account has sufficient credits
+
+**Model Not Available:**
+- Some models may require specific account tiers or permissions
+- Visit the Apertis dashboard to check model availability for your account
+- Try selecting a different model from the `/models` command
+
+**Rate Limiting:**
+- Apertis implements rate limits to ensure fair usage
+- If you encounter rate limit errors, wait a few moments before retrying
+- Consider upgrading your account tier for higher rate limits
+
+---
+
 ### Amazon Bedrock
 
 To use Amazon Bedrock with OpenCode: