{"_id":"@catalystlabs/ai-sdk-provider-meta-llama","name":"@catalystlabs/ai-sdk-provider-meta-llama","dist-tags":{"latest":"0.1.0"},"versions":{"0.1.0":{"name":"@catalystlabs/ai-sdk-provider-meta-llama","version":"0.1.0","description":"Meta Llama 4.0 provider for Vercel AI SDK - Access Llama models through the official Meta API","main":"dist/index.js","types":"dist/index.d.ts","module":"dist/index.mjs","repository":{"type":"git","url":"git+https://github.com/CatalystPracticeLab/ai-sdk-meta-llama.git"},"homepage":"https://github.com/CatalystPracticeLab/ai-sdk-meta-llama#readme","bugs":{"url":"https://github.com/CatalystPracticeLab/ai-sdk-meta-llama/issues"},"scripts":{"build":"tsc","test":"jest","prepublishOnly":"npm run build"},"keywords":["ai","llama","meta","vercel","ai-sdk","llm"],"author":{"name":"Catalyst Labs"},"license":"MIT","dependencies":{"@ai-sdk/provider":"*","@ai-sdk/provider-utils":"*","zod":"^4.1.5"},"devDependencies":{"@types/node":"*","typescript":"*","ai":"*"},"peerDependencies":{"ai":"*"},"_id":"@catalystlabs/ai-sdk-provider-meta-llama@0.1.0","_integrity":"sha512-gChvo4XW2PHKZp+3bbg8GvUbGplazJMCoj9kNTvyhapmI8w0DivIv+qnVAT5oGBi7Z7BFP2RoUjXTqZNB5fETg==","_resolved":"/Users/catalystlabs/local_projects/ai-sdk-meta-llama/catalystlabs-ai-sdk-provider-meta-llama-0.1.0.tgz","_from":"file:catalystlabs-ai-sdk-provider-meta-llama-0.1.0.tgz","_nodeVersion":"24.4.1","_npmVersion":"11.4.2","dist":{"integrity":"sha512-gChvo4XW2PHKZp+3bbg8GvUbGplazJMCoj9kNTvyhapmI8w0DivIv+qnVAT5oGBi7Z7BFP2RoUjXTqZNB5fETg==","shasum":"a8c0f93b124e491a2f3ca5a0c5aee72936618845","tarball":"https://registry.npmjs.org/@catalystlabs/ai-sdk-provider-meta-llama/-/ai-sdk-provider-meta-llama-0.1.0.tgz","fileCount":9,"unpackedSize":449626,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEYCIQDL+nMtDYxiScraujnvkpyJFM7SATAM3J/O6ngWQDR62QIhAL5evNakhSZq9fDHEndF1ZJq+7/zmzGlv0XfFuS+l08D"}]},"_npmUser":{"name":"ccollier86","email":"casey@catalystpractice.com"},"directories":{},"maintainers":[{"name":"ccollier86","email":"casey@catalystpractice.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/ai-sdk-provider-meta-llama_0.1.0_1756649478365_0.7447787989618422"},"_hasShrinkwrap":false}},"time":{"created":"2025-08-31T14:11:18.266Z","0.1.0":"2025-08-31T14:11:18.579Z","modified":"2025-08-31T14:11:18.964Z"},"maintainers":[{"name":"ccollier86","email":"casey@catalystpractice.com"}],"description":"Meta Llama 4.0 provider for Vercel AI SDK - Access Llama models through the official Meta API","homepage":"https://github.com/CatalystPracticeLab/ai-sdk-meta-llama#readme","keywords":["ai","llama","meta","vercel","ai-sdk","llm"],"repository":{"type":"git","url":"git+https://github.com/CatalystPracticeLab/ai-sdk-meta-llama.git"},"author":{"name":"Catalyst Labs"},"bugs":{"url":"https://github.com/CatalystPracticeLab/ai-sdk-meta-llama/issues"},"license":"MIT","readme":"---\r\ntitle: Meta Llama\r\ndescription: Meta Llama Provider for the AI SDK\r\n---\r\n\r\n# Meta Llama\r\n\r\nMeta Llama is the official API for Meta's Llama models, providing access to the latest Llama 4.0 series with advanced capabilities. The Meta Llama provider for the AI SDK enables seamless integration with these powerful models:\r\n\r\n- **Llama 4.0 Models**: Access to Maverick-17B-128E and Scout-17B-16E with mixture-of-experts architecture\r\n- **Multi-Modal Support**: Native image understanding capabilities in Llama 4 models\r\n- **Tool Calling**: Built-in function calling for structured outputs\r\n- **High Performance**: Optimized for both speed and quality\r\n- **Extended Context**: Support for long-form conversations and documents\r\n- **Real-Time Streaming**: Token streaming for responsive applications\r\n\r\nLearn more about Meta Llama's capabilities in the [Meta Llama Documentation](https://llama.developer.meta.com/).\r\n\r\n## Setup\r\n\r\nThe Meta Llama provider is available in the `@catalystlabs/ai-sdk-provider-meta-llama` module. You can install it with:\r\n\r\n<Tabs items={['pnpm', 'npm', 'yarn', 'bun']}>\r\n  <Tab value=\"pnpm\">\r\n    ```bash\r\n    pnpm add @catalystlabs/ai-sdk-provider-meta-llama\r\n    ```\r\n  </Tab>\r\n  <Tab value=\"npm\">\r\n    ```bash\r\n    npm install @catalystlabs/ai-sdk-provider-meta-llama\r\n    ```\r\n  </Tab>\r\n  <Tab value=\"yarn\">\r\n    ```bash\r\n    yarn add @catalystlabs/ai-sdk-provider-meta-llama\r\n    ```\r\n  </Tab>\r\n  <Tab value=\"bun\">\r\n    ```bash\r\n    bun add @catalystlabs/ai-sdk-provider-meta-llama\r\n    ```\r\n  </Tab>\r\n</Tabs>\r\n\r\n## Provider Instance\r\n\r\nTo create a Meta Llama provider instance, use the `createMetaLlama` function:\r\n\r\n```typescript\r\nimport { createMetaLlama } from '@catalystlabs/ai-sdk-provider-meta-llama';\r\n\r\nconst metaLlama = createMetaLlama({\r\n  apiKey: 'YOUR_LLAMA_API_KEY',\r\n});\r\n```\r\n\r\nYou can obtain your Meta Llama API key from the [Meta Llama Dashboard](https://llama.developer.meta.com/).\r\n\r\n## Language Models\r\n\r\nMeta Llama supports chat models. Use the provider instance with model IDs directly:\r\n\r\n```typescript\r\n// Latest Llama 4 model\r\nconst model = metaLlama('Llama-4-Maverick-17B-128E-Instruct-FP8');\r\n\r\n// Efficient variant\r\nconst efficientModel = metaLlama('Llama-4-Scout-17B-16E-Instruct');\r\n```\r\n\r\nYou can find the full list of available models in the [Meta Llama Models documentation](https://llama.developer.meta.com/docs/models).\r\n\r\n## Examples\r\n\r\nHere are examples of using Meta Llama with the AI SDK:\r\n\r\n### generateText\r\n\r\n```typescript\r\nimport { createMetaLlama } from '@catalystlabs/ai-sdk-provider-meta-llama';\r\nimport { generateText } from 'ai';\r\n\r\nconst metaLlama = createMetaLlama({\r\n  apiKey: 'YOUR_LLAMA_API_KEY',\r\n});\r\n\r\nconst { text } = await generateText({\r\n  model: metaLlama('Llama-4-Maverick-17B-128E-Instruct-FP8'),\r\n  prompt: 'What is Meta Llama?',\r\n});\r\n\r\nconsole.log(text);\r\n```\r\n\r\n### streamText\r\n\r\n```typescript\r\nimport { createMetaLlama } from '@catalystlabs/ai-sdk-provider-meta-llama';\r\nimport { streamText } from 'ai';\r\n\r\nconst metaLlama = createMetaLlama({\r\n  apiKey: 'YOUR_LLAMA_API_KEY',\r\n});\r\n\r\nconst result = streamText({\r\n  model: metaLlama('Llama-4-Maverick-17B-128E-Instruct-FP8'),\r\n  prompt: 'Write a short story about AI.',\r\n});\r\n\r\nfor await (const chunk of result.textStream) {\r\n  console.log(chunk);\r\n}\r\n```\r\n\r\n## Advanced Features\r\n\r\nMeta Llama offers several advanced features to enhance your AI applications:\r\n\r\n**Multi-Modal Processing**: Llama 4 models can understand and analyze images alongside text inputs.\r\n\r\n**Tool Calling**: Execute functions and retrieve structured data with built-in tool support.\r\n\r\n**Mixture of Experts**: Llama 4 Maverick uses 128 experts for superior performance across diverse tasks.\r\n\r\n**Token Efficiency**: Optimized tokenization for better context utilization and cost management.\r\n\r\n**Safety Features**: Built-in safety measures and content filtering aligned with Meta's responsible AI principles.\r\n\r\nFor more information about these features and advanced configuration options, visit the [Meta Llama Documentation](https://llama.developer.meta.com/).\r\n\r\n## Additional Resources\r\n\r\n- [Meta Llama Provider Repository](https://github.com/CatalystPracticeLab/ai-sdk-meta-llama)\r\n- [Meta Llama Documentation](https://llama.developer.meta.com/)\r\n- [Meta Llama Dashboard](https://llama.developer.meta.com/)\r\n- [Meta AI Research](https://ai.meta.com/llama/)\r\n- [Llama Stack GitHub](https://github.com/meta-llama/llama-stack)","readmeFilename":"README.md","_rev":"1-5b6764ead225d20997496044e40f2fce"}