Add SLNG model provider

core/llm/llms/Slng.ts

@@ -0,0 +1,163 @@
+/**
+ * SLNG Provider for Continue.dev
+ *
+ * This provider enables access to 19 SLNG voice AI models (11 TTS + 8 STT)
+ * across EU, AU, and Asia regions.
+ *
+ * Documentation: https://docs.slng.ai
+ * Models: https://docs.slng.ai/models
+ */
+
+import { LLM } from "../../index.js";
+import { CompletionOptions, LLMOptions, ModelProvider } from "../../index.js";
+
+class Slng implements LLM {
+  static providerName: ModelProvider = "slng";
+  static defaultOptions: Partial<LLMOptions> = {
+    apiBase: "https://api.slng.ai/v1",
+  };
+
+  private apiKey: string;
+  private apiBase: string;
+
+  constructor(options: LLMOptions) {
+    this.apiKey = options.apiKey || "";
+    this.apiBase = options.apiBase || "https://api.slng.ai/v1";
+
+    if (!this.apiKey) {
+      throw new Error(
+        "SLNG API key is required. Set it in your Continue config or via SLNG_API_KEY environment variable."
+      );
+    }
+  }
+
+  async *streamComplete(
+    prompt: string,
+    options: CompletionOptions
+  ): AsyncGenerator<string> {
+    const endpoint = this.getEndpoint(options.model);
+
+    const response = await fetch(`${this.apiBase}${endpoint}`, {
+      method: "POST",
+      headers: {
+        Authorization: `Bearer ${this.apiKey}`,
+        "Content-Type": "application/json",
+      },
+      body: JSON.stringify({
+        input: prompt,
+        model: options.model,
+        stream: true,
+        temperature: options.temperature,
+        max_tokens: options.maxTokens,
+      }),
+    });
+
+    if (!response.ok) {
+      throw new Error(
+        `SLNG API error: ${response.status} ${response.statusText}`
+      );
+    }
+
+    const reader = response.body?.getReader();
+    if (!reader) {
+      throw new Error("No response body");
+    }
+
+    const decoder = new TextDecoder();
+
+    while (true) {
+      const { done, value } = await reader.read();
+      if (done) break;
+
+      const chunk = decoder.decode(value);
+      const lines = chunk.split("\n").filter((line) => line.trim() !== "");
+
+      for (const line of lines) {
+        if (line.startsWith("data: ")) {
+          const data = line.slice(6);
+          if (data === "[DONE]") continue;
+
+          try {
+            const parsed = JSON.parse(data);
+            const text = parsed.text || parsed.delta?.text || "";
+            if (text) yield text;
+          } catch (e) {
+            // Skip invalid JSON
+            console.warn("Failed to parse SSE data:", data);
+          }
+        }
+      }
+    }
+  }
+
+  async complete(prompt: string, options: CompletionOptions): Promise<string> {
+    const endpoint = this.getEndpoint(options.model);
+
+    const response = await fetch(`${this.apiBase}${endpoint}`, {
+      method: "POST",
+      headers: {
+        Authorization: `Bearer ${this.apiKey}`,
+        "Content-Type": "application/json",
+      },
+      body: JSON.stringify({
+        input: prompt,
+        model: options.model,
+        temperature: options.temperature,
+        max_tokens: options.maxTokens,
+      }),
+    });
+
+    if (!response.ok) {
+      throw new Error(
+        `SLNG API error: ${response.status} ${response.statusText}`
+      );
+    }
+
+    const data = await response.json();
+    return data.text || data.transcript || "";
+  }
+
+  listModels(): string[] {
+    return [
+      // TTS Models (11)
+      "deepgram-aura-2",
+      "bulbul-v3",
+      "cartesia-sonic-3",
+      "cartesia-sonic-3-5",
+      "fish-tts-s2-1-pro",
+      "gradium-tts",
+      "inworld-max-1-5",
+      "kugel-2",
+      "murf-falcon",
+      "sarvam-saaras",
+      "soniox-tts-rt-v1",
+      // STT Models (8)
+      "deepgram-nova-3",
+      "deepgram-nova-3-medical",
+      "deepgram-nova-3-multi",
+      "fish-audio-asr",
+      "gradium-stt",
+      "reson8-stt-v1",
+      "soniox-speech-ai-rt-v5",
+      "speechmatics-realtime-v2",
+    ];
+  }
+
+  private getEndpoint(model: string): string {
+    // Determine if TTS or STT based on model name
+    const sttKeywords = ["nova", "asr", "stt", "speechmatics", "soniox-speech"];
+    const isSTT = sttKeywords.some((keyword) => model.includes(keyword));
+
+    if (isSTT) {
+      return `/stt/slng/${model}`;
+    } else {
+      return `/tts/slng/${model}`;
+    }
+  }
+
+  supportsFim(): boolean {
+    return false;
+  }
+}
+
+export default Slng;

docs/customize/model-providers/more/slng.mdx

@@ -0,0 +1,114 @@
+# SLNG
+
+SLNG is a regional voice AI infrastructure provider offering TTS and STT models across EU, AU, and Asia with sub-100ms latency.
+
+## Configuration
+
+Add SLNG to your `~/.continue/config.json`:
+
+```json
+{
+  "models": [
+    {
+      "title": "SLNG - Deepgram Aura 2",
+      "provider": "slng",
+      "model": "deepgram-aura-2",
+      "apiKey": "your-slng-api-key"
+    }
+  ]
+}
+```
+
+You can also set the API key via environment variable:
+```bash
+export SLNG_API_KEY="your-slng-api-key"
+```
+
+## Available Models
+
+### TTS (Text-to-Speech) - 11 Models
+
+| Model ID | Name | Languages | Use Case |
+|----------|------|-----------|----------|
+| `deepgram-aura-2` | Deepgram Aura 2 | English, Spanish | Natural voices |
+| `bulbul-v3` | Sarvam Bulbul v3 | 23 Indian languages | Regional voices |
+| `cartesia-sonic-3` | Cartesia Sonic 3 | Multilingual | Low latency |
+| `cartesia-sonic-3-5` | Cartesia Sonic 3.5 | Multilingual | Enhanced quality |
+| `fish-tts-s2-1-pro` | Fish TTS S2.1 Pro | EN, ZH, JA | Asian languages |
+| `gradium-tts` | Gradium TTS | English | Regional deployment |
+| `inworld-max-1-5` | Inworld Max 1.5 | English | Character voices |
+| `kugel-2` | KugelAudio Kugel 2 | European languages | EU deployment |
+| `murf-falcon` | Murf Falcon | English | Professional narration |
+| `sarvam-saaras` | Sarvam AI Saaras | Indian languages | Regional voices |
+| `soniox-tts-rt-v1` | Soniox TTS RT v1 | English | Real-time synthesis |
+
+### STT (Speech-to-Text) - 8 Models
+
+| Model ID | Name | Languages | Use Case |
+|----------|------|-----------|----------|
+| `deepgram-nova-3` | Deepgram Nova 3 | English | Standard transcription |
+| `deepgram-nova-3-medical` | Deepgram Nova 3 Medical | English | Medical terminology |
+| `deepgram-nova-3-multi` | Deepgram Nova 3 Multi | EN, ES, HI | Multilingual |
+| `fish-audio-asr` | Fish Audio ASR | EN, ZH, JA | Asian languages |
+| `gradium-stt` | Gradium STT | English | Regional deployment |
+| `reson8-stt-v1` | Reson8 STT v1 | English | Enterprise-grade |
+| `soniox-speech-ai-rt-v5` | Soniox Speech AI RT v5 | English | High accuracy |
+| `speechmatics-realtime-v2` | Speechmatics Realtime v2 | Multilingual | Real-time |
+
+## Regional Deployment
+
+SLNG models are deployed across multiple regions:
+- **US:** Central US infrastructure
+- **EU:** European data centers
+- **AU:** Australia deployment
+- **Asia:** India, Singapore, and more
+
+Models automatically route to the nearest region for optimal latency.
+
+## Authentication
+
+Get your API key from [SLNG](https://app.slng.ai/sign-in).
+
+## Example Configurations
+
+### TTS for Documentation
+```json
+{
+  "title": "SLNG TTS - Aura 2",
+  "provider": "slng",
+  "model": "deepgram-aura-2",
+  "apiKey": "your-api-key"
+}
+```
+
+### STT for Transcription
+```json
+{
+  "title": "SLNG STT - Nova 3",
+  "provider": "slng",
+  "model": "deepgram-nova-3",
+  "apiKey": "your-api-key"
+}
+```
+
+### Multilingual Support
+```json
+{
+  "title": "SLNG - Indian Languages",
+  "provider": "slng",
+  "model": "bulbul-v3",
+  "apiKey": "your-api-key"
+}
+```
+
+## Documentation
+
+- [SLNG API Documentation](https://docs.slng.ai)
+- [Model Catalog](https://docs.slng.ai/models)
+- [API Reference](https://docs.slng.ai/api-reference)
+
+## Support
+
+For issues or questions:
+- GitHub: [continuedev/continue](https://github.com/continuedev/continue/issues)
+- SLNG Docs: https://docs.slng.ai