feat(ai-nebius): add Nebius Token Factory adapter

docs/adapters/nebius.md

@@ -0,0 +1,271 @@
+---
+title: Nebius Token Factory
+id: nebius-adapter
+order: 5
+---
+
+The Nebius Token Factory adapter provides access to various LLMs hosted on Nebius Token Factory, including DeepSeek, Llama, Qwen, and more. Nebius Token Factory offers an OpenAI-compatible API, making it easy to use these models with TanStack AI.
+
+## Installation
+
+```bash
+npm install @tanstack/ai-nebius
+```
+
+## Basic Usage
+
+```typescript
+import { chat } from "@tanstack/ai";
+import { nebiusText } from "@tanstack/ai-nebius";
+
+const stream = chat({
+  adapter: nebiusText("deepseek-ai/DeepSeek-R1-0528"),
+  messages: [{ role: "user", content: "Hello!" }],
+});
+```
+
+## Basic Usage - Custom API Key
+
+```typescript
+import { chat } from "@tanstack/ai";
+import { createNebiusChat } from "@tanstack/ai-nebius";
+
+const adapter = createNebiusChat(
+  "deepseek-ai/DeepSeek-R1-0528",
+  "your-api-key-here"
+);
+
+const stream = chat({
+  adapter,
+  messages: [{ role: "user", content: "Hello!" }],
+});
+```
+
+## Configuration
+
+```typescript
+import { createNebiusChat } from "@tanstack/ai-nebius";
+
+// With explicit API key
+const adapter = createNebiusChat("deepseek-ai/DeepSeek-R1-0528", "your-api-key");
+
+// With custom base URL (optional, defaults to Nebius endpoint)
+const adapter = createNebiusChat(
+  "deepseek-ai/DeepSeek-R1-0528",
+  "your-api-key",
+  { baseURL: "https://api.tokenfactory.nebius.com/v1/" }
+);
+```
+
+## Available Models
+
+Nebius Token Factory offers a variety of models. You can use any model name as a string, but here are some popular options:
+
+### DeepSeek Models
+- `deepseek-ai/DeepSeek-R1-0528` - DeepSeek R1 reasoning model
+- `deepseek-ai/DeepSeek-V3-0324` - DeepSeek V3 model
+
+### Llama Models
+- `meta-llama/Meta-Llama-3.1-70B-Instruct` - Llama 3.1 70B
+- `meta-llama/Meta-Llama-3.1-8B-Instruct` - Llama 3.1 8B
+- `meta-llama/Meta-Llama-3.1-405B-Instruct` - Llama 3.1 405B
+
+### Qwen Models
+- `Qwen/Qwen2.5-72B-Instruct` - Qwen 2.5 72B
+- `Qwen/Qwen2.5-7B-Instruct` - Qwen 2.5 7B
+- `Qwen/Qwen2.5-32B-Instruct` - Qwen 2.5 32B
+
+To see all available models, check the [Nebius Token Factory documentation](https://docs.tokenfactory.nebius.com/ai-models-inference/overview).
+
+## Example: Chat Completion
+
+```typescript
+import { chat, toStreamResponse } from "@tanstack/ai";
+import { nebiusText } from "@tanstack/ai-nebius";
+
+export async function POST(request: Request) {
+  const { messages } = await request.json();
+
+  const stream = chat({
+    adapter: nebiusText("deepseek-ai/DeepSeek-R1-0528"),
+    messages,
+  });
+
+  return toStreamResponse(stream);
+}
+```
+
+## Example: With Tools
+
+```typescript
+import { chat, toolDefinition } from "@tanstack/ai";
+import { nebiusText } from "@tanstack/ai-nebius";
+import { z } from "zod";
+
+const getWeatherDef = toolDefinition({
+  name: "get_weather",
+  description: "Get the current weather",
+  inputSchema: z.object({
+    location: z.string(),
+  }),
+});
+
+const getWeather = getWeatherDef.server(async ({ location }) => {
+  // Fetch weather data
+  return { temperature: 72, conditions: "sunny" };
+});
+
+const stream = chat({
+  adapter: nebiusText("deepseek-ai/DeepSeek-R1-0528"),
+  messages,
+  tools: [getWeather],
+});
+```
+
+**Note:** Tool support varies by model. Models like DeepSeek R1, Llama 3.1, and Qwen 2.5 generally have good tool calling support.
+
+## Model Options
+
+Nebius supports various provider-specific options compatible with OpenAI's chat completions API:
+
+```typescript
+const stream = chat({
+  adapter: nebiusText("deepseek-ai/DeepSeek-R1-0528"),
+  messages,
+  modelOptions: {
+    temperature: 0.7,
+    top_p: 0.9,
+    max_tokens: 1000,
+    presence_penalty: 0.5,
+    frequency_penalty: 0.5,
+    stop: ["END"],
+  },
+});
+```
+
+### Available Options
+
+```typescript
+modelOptions: {
+  // Sampling
+  temperature: 0.7,        // 0-2, controls randomness
+  top_p: 0.9,              // 0-1, nucleus sampling
+
+  // Generation
+  max_tokens: 1000,        // Maximum tokens to generate
+  stop: ["END"],           // Stop sequences (string or array)
+
+  // Penalties
+  presence_penalty: 0.5,   // -2.0 to 2.0
+  frequency_penalty: 0.5,  // -2.0 to 2.0
+
+  // Advanced
+  logit_bias: {},          // Logit bias map
+  user: "user-123",        // User identifier
+}
+```
+
+## Summarization
+
+Summarize long text content:
+
+```typescript
+import { summarize } from "@tanstack/ai";
+import { nebiusSummarize } from "@tanstack/ai-nebius";
+
+const result = await summarize({
+  adapter: nebiusSummarize("deepseek-ai/DeepSeek-R1-0528"),
+  text: "Your long text to summarize...",
+  maxLength: 100,
+  style: "concise", // "concise" | "bullet-points" | "paragraph"
+});
+
+console.log(result.summary);
+```
+
+## Getting Started with Nebius Token Factory
+
+### 1. Create an Account
+
+Visit [Nebius Token Factory](https://tokenfactory.nebius.com) and sign up using your Google or GitHub account.
+
+### 2. Get an API Key
+
+- Navigate to the **API keys** section in your dashboard
+- Click on **Create API key**
+- Provide a name for the key
+- Save the generated API key securely
+
+### 3. Set Environment Variable
+
+```bash
+NEBIUS_API_KEY=your-api-key-here
+```
+
+## Environment Variables
+
+Optionally set the API key in environment variables:
+
+```bash
+NEBIUS_API_KEY=your-api-key-here
+```
+
+The adapter will automatically detect `NEBIUS_API_KEY` from:
+- `process.env` (Node.js)
+- `window.env` (Browser with injected env)
+
+## API Reference
+
+### `nebiusText(model, config?)`
+
+Creates a Nebius text/chat adapter using environment variables.
+
+**Parameters:**
+- `model` - The model name (e.g., 'deepseek-ai/DeepSeek-R1-0528')
+- `config.apiKey?` - Optional API key (if not provided, uses NEBIUS_API_KEY env var)
+- `config.baseURL?` - Optional base URL (defaults to Nebius endpoint)
+
+**Returns:** A Nebius text adapter instance.
+
+### `createNebiusChat(model, apiKey, config?)`
+
+Creates a Nebius text/chat adapter with an explicit API key.
+
+**Parameters:**
+- `model` - The model name
+- `apiKey` - Your Nebius API key
+- `config.baseURL?` - Optional base URL
+
+**Returns:** A Nebius text adapter instance.
+
+### `nebiusSummarize(model, config?)`
+
+Creates a Nebius summarization adapter using environment variables.
+
+**Returns:** A Nebius summarize adapter instance.
+
+### `createNebiusSummarize(model, apiKey, config?)`
+
+Creates a Nebius summarization adapter with an explicit API key.
+
+**Returns:** A Nebius summarize adapter instance.
+
+## Benefits of Nebius Token Factory
+
+- ✅ **Multiple Models** - Access to DeepSeek, Llama, Qwen, and more
+- ✅ **OpenAI-Compatible** - Familiar API interface
+- ✅ **Cost-Effective** - Competitive pricing for inference
+- ✅ **Scalable** - Enterprise-grade infrastructure
+- ✅ **Easy Integration** - Drop-in replacement for OpenAI-compatible workflows
+
+## Limitations
+
+- **Image Generation**: Nebius Token Factory does not support image generation. Use OpenAI or Gemini for image generation.
+- **Model Availability**: Available models may vary. Check the [Nebius documentation](https://docs.tokenfactory.nebius.com/ai-models-inference/overview) for current offerings.
+
+## Next Steps
+
+- [Getting Started](../getting-started/quick-start) - Learn the basics
+- [Tools Guide](../guides/tools) - Learn about tools
+- [Other Adapters](./openai) - Explore other providers
+

docs/config.json

@@ -139,6 +139,10 @@
         {
           "label": "Ollama",
           "to": "adapters/ollama"
+        },
+        {
+          "label": "Nebius Token Factory",
+          "to": "adapters/nebius"
         }
       ]
     },

knip.json

@@ -27,6 +27,14 @@
     "packages/typescript/ai-openai": {
       "ignore": ["src/tools/**"]
     },
+    "packages/typescript/ai-solid": {
+      "tsdown": false,
+      "ignore": ["tsdown.config.ts"]
+    },
+    "packages/typescript/ai-vue": {
+      "tsdown": false,
+      "ignore": ["tsdown.config.ts"]
+    },
     "packages/typescript/ai-vue-ui": {
       "ignoreDependencies": [
         "@crazydos/vue-markdown",

package.json

@@ -33,7 +33,7 @@
     "format": "prettier --experimental-cli --ignore-unknown '**/*' --write",
     "generate-docs": "node scripts/generate-docs.ts && pnpm run copy:readme",
     "sync-docs-config": "node scripts/sync-docs-config.ts",
-    "copy:readme": "cp README.md packages/typescript/ai/README.md && cp README.md packages/typescript/ai-devtools/README.md && cp README.md packages/typescript/ai-client/README.md && cp README.md packages/typescript/ai-gemini/README.md && cp README.md packages/typescript/ai-ollama/README.md && cp README.md packages/typescript/ai-openai/README.md && cp README.md packages/typescript/ai-react/README.md && cp README.md packages/typescript/ai-react-ui/README.md && cp README.md packages/typescript/react-ai-devtools/README.md && cp README.md packages/typescript/solid-ai-devtools/README.md",
+    "copy:readme": "cp README.md packages/typescript/ai/README.md && cp README.md packages/typescript/ai-devtools/README.md && cp README.md packages/typescript/ai-client/README.md && cp README.md packages/typescript/ai-gemini/README.md && cp README.md packages/typescript/ai-ollama/README.md && cp README.md packages/typescript/ai-openai/README.md && cp README.md packages/typescript/ai-react/README.md && cp README.md packages/typescript/ai-react-ui/README.md && cp README.md packages/typescript/react-ai-devtools/README.md && cp README.md packages/typescript/solid-ai-devtools/README.md ",
     "changeset": "changeset",
     "changeset:publish": "changeset publish",
     "changeset:version": "changeset version && pnpm install --no-frozen-lockfile && pnpm format"

packages/typescript/ai-nebius/CHANGELOG.md

@@ -0,0 +1,12 @@
+# @tanstack/ai-nebius
+
+## 0.2.0
+
+### Minor Changes
+
+- Initial release of Nebius Token Factory adapter for TanStack AI
+  - Text/chat adapter with streaming support
+  - Summarization adapter
+  - Tool/function calling support
+  - OpenAI-compatible API using `https://api.tokenfactory.nebius.com/v1/`
+  - Support for DeepSeek, Llama, Qwen models

packages/typescript/ai-nebius/README.md

@@ -0,0 +1,25 @@
+# @tanstack/ai-nebius
+
+Nebius Token Factory adapter for TanStack AI.
+
+## Installation
+
+```bash
+npm install @tanstack/ai-nebius
+```
+
+## Usage
+
+```typescript
+import { chat } from '@tanstack/ai'
+import { nebiusText } from '@tanstack/ai-nebius'
+
+const stream = chat({
+  adapter: nebiusText('deepseek-ai/DeepSeek-R1-0528'),
+  messages: [{ role: 'user', content: 'Hello!' }],
+})
+```
+
+## Documentation
+
+See the [Nebius Token Factory adapter documentation](../../../docs/adapters/nebius.md) for full details.