hatchet/frontend/docs/components/LLMIntegrationTabs.mdx

import { Callout, Tabs } from "nextra/components";
import { snippets } from "@/lib/generated/snippets";
import { Snippet } from "@/components/code";
import PackageManagerInstall from "@/components/PackageManagerInstall";
import UniversalTabs from "@/components/UniversalTabs";

<Tabs items={["OpenAI", "Anthropic", "Groq", "Vercel AI SDK", "Ollama"]}>
  <Tabs.Tab title="OpenAI">

OpenAI's [Chat Completions API](https://platform.openai.com/docs/guides/text-generation) provides access to GPT models for text generation, function calling, and structured outputs. It's the most widely adopted LLM API and supports streaming, tool use, and JSON mode.

  <UniversalTabs items={["Python", "TypeScript", "Go", "Ruby"]} variant="hidden">
    <Tabs.Tab title="Python">
      <PackageManagerInstall packages={{ python: "openai" }} />
      <Snippet src={snippets.python.guides.integrations.llm_openai.open_ai_usage} />
    </Tabs.Tab>
    <Tabs.Tab title="TypeScript">
      <PackageManagerInstall packages={{ typescript: "openai" }} />
      <Snippet src={snippets.typescript.guides.integrations.llm_openai.open_ai_usage} />
    </Tabs.Tab>
    <Tabs.Tab title="Go">
      <PackageManagerInstall packages={{ go: "github.com/sashabaranov/go-openai" }} />
      <Snippet src={snippets.go.guides.integrations.llm_openai.open_ai_usage} />
    </Tabs.Tab>
    <Tabs.Tab title="Ruby">
      <PackageManagerInstall packages={{ ruby: "openai" }} />
      <Snippet src={snippets.ruby.guides.integrations.llm_openai.open_ai_usage} />
    </Tabs.Tab>
  </UniversalTabs>
  </Tabs.Tab>
  <Tabs.Tab title="Anthropic">

Anthropic's [Messages API](https://docs.anthropic.com/en/docs/build-with-claude/text-generation) powers the Claude family of models, including Claude Sonnet and Claude Haiku. Claude excels at long-context reasoning, careful instruction following, and tool use with extended thinking support.

  <UniversalTabs items={["Python", "TypeScript", "Go", "Ruby"]} variant="hidden">
    <Tabs.Tab title="Python">
      <PackageManagerInstall packages={{ python: "anthropic" }} />
      <Snippet src={snippets.python.guides.integrations.llm_anthropic.anthropic_usage} />
    </Tabs.Tab>
    <Tabs.Tab title="TypeScript">
      <PackageManagerInstall packages={{ typescript: "@anthropic-ai/sdk" }} />
      <Snippet src={snippets.typescript.guides.integrations.llm_anthropic.anthropic_usage} />
    </Tabs.Tab>
    <Tabs.Tab title="Go">
      <Callout type="info">
        Anthropic Go: `go get github.com/anthropics/anthropic-sdk-go`, wire `messages.Create()` into your complete function.
      </Callout>
    </Tabs.Tab>
    <Tabs.Tab title="Ruby">
      <Callout type="info">
        Anthropic Ruby: `bundle add anthropic`, wire the client into your complete function.
      </Callout>
    </Tabs.Tab>
  </UniversalTabs>
  </Tabs.Tab>
  <Tabs.Tab title="Groq">

[Groq](https://console.groq.com/docs/overview) provides ultra-fast inference for open-source models like Llama and Mixtral using custom LPU hardware. Its OpenAI-compatible API makes it a drop-in replacement when you need low latency.

  <UniversalTabs items={["Python", "TypeScript", "Go", "Ruby"]} variant="hidden">
    <Tabs.Tab title="Python">
      <PackageManagerInstall packages={{ python: "groq" }} />
      <Snippet src={snippets.python.guides.integrations.llm_groq.groq_usage} />
    </Tabs.Tab>
    <Tabs.Tab title="TypeScript">
      <PackageManagerInstall packages={{ typescript: "groq-sdk" }} />
      <Snippet src={snippets.typescript.guides.integrations.llm_groq.groq_usage} />
    </Tabs.Tab>
    <Tabs.Tab title="Go">
      <Callout type="info">
        Groq: use `net/http` against `api.groq.com/openai/v1/chat/completions`. See [Groq docs](https://console.groq.com/docs).
      </Callout>
    </Tabs.Tab>
    <Tabs.Tab title="Ruby">
      <Callout type="info">
        Groq Ruby: `bundle add groq` or use HTTP client. See [Groq docs](https://console.groq.com/docs).
      </Callout>
    </Tabs.Tab>
  </UniversalTabs>
  </Tabs.Tab>
  <Tabs.Tab title="Vercel AI SDK">

The [Vercel AI SDK](https://sdk.vercel.ai/docs) is a TypeScript toolkit that provides a unified interface across providers (OpenAI, Anthropic, Google, and more). It includes helpers for streaming, tool calls, and structured object generation via `generateText` and `streamText`.

  <UniversalTabs items={["Python", "TypeScript", "Go", "Ruby"]} variant="hidden">
    <Tabs.Tab title="Python">
      <Callout type="info">
        Vercel AI SDK is JavaScript/TypeScript only. Use OpenAI, Anthropic, or Groq SDK directly.
      </Callout>
    </Tabs.Tab>
    <Tabs.Tab title="TypeScript">
      <PackageManagerInstall packages={{ typescript: "ai @ai-sdk/openai" }} />
      <Snippet src={snippets.typescript.guides.integrations.llm_vercel_ai_sdk.vercel_ai_sdk_usage} />
    </Tabs.Tab>
    <Tabs.Tab title="Go">
      <Callout type="info">
        Vercel AI SDK is JavaScript/TypeScript only.
      </Callout>
    </Tabs.Tab>
    <Tabs.Tab title="Ruby">
      <Callout type="info">
        Vercel AI SDK is JavaScript/TypeScript only.
      </Callout>
    </Tabs.Tab>
  </UniversalTabs>
  </Tabs.Tab>
  <Tabs.Tab title="Ollama">

[Ollama](https://ollama.com/) runs open-source models locally with no API key required. It supports Llama, Mistral, Gemma, and others through a simple REST API on `localhost:11434`. Ideal for development, air-gapped environments, or when you want full control over your model.

  <UniversalTabs items={["Python", "TypeScript", "Go", "Ruby"]} variant="hidden">
    <Tabs.Tab title="Python">
      <PackageManagerInstall packages={{ python: "ollama" }} />
      <Snippet src={snippets.python.guides.integrations.llm_ollama.ollama_usage} />
    </Tabs.Tab>
    <Tabs.Tab title="TypeScript">
      <Callout type="info">
        Use `fetch` to `http://localhost:11434/api/chat`. See [Ollama API](https://github.com/ollama/ollama/blob/main/docs/api.md).
      </Callout>
    </Tabs.Tab>
    <Tabs.Tab title="Go">
      <Callout type="info">
        Use `net/http` to `http://localhost:11434/api/chat`. See [Ollama API](https://github.com/ollama/ollama/blob/main/docs/api.md).
      </Callout>
    </Tabs.Tab>
    <Tabs.Tab title="Ruby">
      <Callout type="info">
        Use `Net::HTTP` to `http://localhost:11434/api/chat`. See [Ollama API](https://github.com/ollama/ollama/blob/main/docs/api.md).
      </Callout>
    </Tabs.Tab>
  </UniversalTabs>
  </Tabs.Tab>
</Tabs>