liter_llm 1.0.0.pre.rc.6-aarch64-linux

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: 2837ee8e05be49794fa8b1e1885b2a8149584ab2d79de3d98c7bec95cbdaf978
+  data.tar.gz: 4ef79499547e416da331335556cdcfbaec7502dd7ebbaf5be5a4207b5072b77e
+SHA512:
+  metadata.gz: 1dbdd75dd67268d3f69c543ef5ac97d7f539f1976d65f44838b83f7e2213dd87611e74b5eb08c30cb86ad59688086438878701f97e3483ea9b170d0c209bc1d2
+  data.tar.gz: 351ddc8832eb89dc60d09a9c1f23f46ed85eeb5885b072b125ab076baaa65cf1f1b7a03981493c48fd491a21554acbf560c468157ccade978a63da8365abbea3

data/README.md

@@ -0,0 +1,239 @@
+# Ruby
+
+<div align="center" style="display: flex; flex-wrap: wrap; gap: 8px; justify-content: center; margin: 20px 0;">
+  <!-- Language Bindings -->
+  <a href="https://crates.io/crates/liter-llm">
+    <img src="https://img.shields.io/crates/v/liter-llm?label=Rust&color=007ec6" alt="Rust">
+  </a>
+  <a href="https://pypi.org/project/liter-llm/">
+    <img src="https://img.shields.io/pypi/v/liter-llm?label=Python&color=007ec6" alt="Python">
+  </a>
+  <a href="https://www.npmjs.com/package/@kreuzberg/liter-llm">
+    <img src="https://img.shields.io/npm/v/@kreuzberg/liter-llm?label=Node.js&color=007ec6" alt="Node.js">
+  </a>
+  <a href="https://www.npmjs.com/package/@kreuzberg/liter-llm-wasm">
+    <img src="https://img.shields.io/npm/v/@kreuzberg/liter-llm-wasm?label=WASM&color=007ec6" alt="WASM">
+  </a>
+  <a href="https://central.sonatype.com/artifact/dev.kreuzberg/liter-llm">
+    <img src="https://img.shields.io/maven-central/v/dev.kreuzberg/liter-llm?label=Java&color=007ec6" alt="Java">
+  </a>
+  <a href="https://github.com/kreuzberg-dev/liter-llm/tree/main/packages/go">
+    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/liter-llm?label=Go&color=007ec6" alt="Go">
+  </a>
+  <a href="https://www.nuget.org/packages/LiterLlm">
+    <img src="https://img.shields.io/nuget/v/LiterLlm?label=C%23&color=007ec6" alt="C#">
+  </a>
+  <a href="https://packagist.org/packages/kreuzberg/liter-llm">
+    <img src="https://img.shields.io/packagist/v/kreuzberg/liter-llm?label=PHP&color=007ec6" alt="PHP">
+  </a>
+  <a href="https://rubygems.org/gems/liter_llm">
+    <img src="https://img.shields.io/gem/v/liter_llm?label=Ruby&color=007ec6" alt="Ruby">
+  </a>
+  <a href="https://hex.pm/packages/liter_llm">
+    <img src="https://img.shields.io/hexpm/v/liter_llm?label=Elixir&color=007ec6" alt="Elixir">
+  </a>
+  <a href="https://github.com/kreuzberg-dev/liter-llm/pkgs/container/liter-llm">
+    <img src="https://img.shields.io/badge/Docker-007ec6?logo=docker&logoColor=white" alt="Docker">
+  </a>
+  <a href="https://github.com/kreuzberg-dev/liter-llm/tree/main/crates/liter-llm-ffi">
+    <img src="https://img.shields.io/badge/C-FFI-007ec6" alt="C FFI">
+  </a>
+
+  <!-- Project Info -->
+  <a href="https://github.com/kreuzberg-dev/liter-llm/blob/main/LICENSE">
+    <img src="https://img.shields.io/badge/License-MIT-007ec6" alt="License">
+  </a>
+  <a href="https://docs.liter-llm.kreuzberg.dev">
+    <img src="https://img.shields.io/badge/docs-kreuzberg.dev-007ec6" alt="Docs">
+  </a>
+</div>
+
+
+<div align="center" style="margin: 20px 0;">
+  <picture>
+    <img width="100%" alt="kreuzberg.dev" src="https://github.com/user-attachments/assets/1b6c6ad7-3b6d-4171-b1c9-f2026cc9deb8" />
+  </picture>
+</div>
+
+
+<div align="center" style="margin-bottom: 20px;">
+  <a href="https://discord.gg/xt9WY3GnKR">
+    <img height="22" src="https://img.shields.io/badge/Discord-Join%20our%20community-7289da?logo=discord&logoColor=white" alt="Discord">
+  </a>
+</div>
+
+
+Universal LLM API client for Ruby. Access 142+ LLM providers through a single interface with idiomatic Ruby API and native performance.
+
+
+## Installation
+
+### Package Installation
+
+
+Install via one of the supported package managers:
+
+
+
+**gem:**
+
+```bash
+gem install liter_llm
+```
+
+
+
+
+**Bundler:**
+
+```ruby
+gem 'liter_llm'
+```
+
+
+
+
+
+### System Requirements
+
+
+- **Ruby 3.2+** required
+- API keys via environment variables (e.g. `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`)
+
+
+
+## Quick Start
+
+### Basic Chat
+
+Send a message to any provider using the `provider/model` prefix:
+
+```ruby
+# frozen_string_literal: true
+
+require "liter_llm"
+require "json"
+
+client = LiterLlm::LlmClient.new(ENV.fetch("OPENAI_API_KEY"), {})
+
+response = JSON.parse(client.chat(JSON.generate(
+  model: "openai/gpt-4o",
+  messages: [{ role: "user", content: "Hello!" }]
+)))
+
+puts response.dig("choices", 0, "message", "content")
+```
+
+
+### Common Use Cases
+
+
+#### Streaming Responses
+
+Stream tokens in real time:
+
+```ruby
+# frozen_string_literal: true
+
+require "liter_llm"
+require "json"
+
+client = LiterLlm::LlmClient.new(ENV.fetch("OPENAI_API_KEY"), {})
+
+chunks = JSON.parse(client.chat_stream(JSON.generate(
+  model: "openai/gpt-4o-mini",
+  messages: [{ role: "user", content: "Hello" }]
+)))
+
+chunks.each { |chunk| puts chunk }
+```
+
+
+
+
+
+### Next Steps
+
+- **[Provider Registry](https://github.com/kreuzberg-dev/liter-llm/blob/main/schemas/providers.json)** - Full list of supported providers
+- **[GitHub Repository](https://github.com/kreuzberg-dev/liter-llm)** - Source, issues, and discussions
+
+
+
+## Features
+
+### Supported Providers (142+)
+
+Route to any provider using the `provider/model` prefix convention:
+
+| Provider | Example Model |
+|----------|--------------|
+| **OpenAI** | `openai/gpt-4o`, `openai/gpt-4o-mini` |
+| **Anthropic** | `anthropic/claude-3-5-sonnet-20241022` |
+| **Groq** | `groq/llama-3.1-70b-versatile` |
+| **Mistral** | `mistral/mistral-large-latest` |
+| **Cohere** | `cohere/command-r-plus` |
+| **Together AI** | `together/meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo` |
+| **Fireworks** | `fireworks/accounts/fireworks/models/llama-v3p1-70b-instruct` |
+| **Google Vertex** | `vertexai/gemini-1.5-pro` |
+| **Amazon Bedrock** | `bedrock/anthropic.claude-3-5-sonnet-20241022-v2:0` |
+
+**[Complete Provider List](https://github.com/kreuzberg-dev/liter-llm/blob/main/schemas/providers.json)**
+
+### Key Capabilities
+
+- **Provider Routing** -- Single client for 142+ LLM providers via `provider/model` prefix
+- **Unified API** -- Consistent `chat`, `chat_stream`, `embeddings`, `list_models` interface
+
+- **Streaming** -- Real-time token streaming via `chat_stream`
+
+
+- **Tool Calling** -- Function calling and tool use across all supporting providers
+
+- **Type Safe** -- Schema-driven types compiled from JSON schemas
+- **Secure** -- API keys never logged or serialized, managed via environment variables
+- **Observability** -- Built-in [OpenTelemetry](https://opentelemetry.io/docs/specs/semconv/gen-ai/) with GenAI semantic conventions
+- **Error Handling** -- Structured errors with provider context and retry hints
+
+### Performance
+
+Built on a compiled Rust core for speed and safety:
+
+- **Provider resolution** at client construction -- zero per-request overhead
+- **Configurable timeouts** and connection pooling
+- **Zero-copy streaming** with SSE and AWS EventStream support
+- **API keys** wrapped in secure memory, zeroed on drop
+
+
+
+## Provider Routing
+
+Route to 142+ providers using the `provider/model` prefix convention:
+
+```text
+openai/gpt-4o
+anthropic/claude-3-5-sonnet-20241022
+groq/llama-3.1-70b-versatile
+mistral/mistral-large-latest
+```
+
+See the [provider registry](https://github.com/kreuzberg-dev/liter-llm/blob/main/schemas/providers.json) for the full list.
+
+
+
+## Documentation
+
+- **[Documentation](https://docs.liter-llm.kreuzberg.dev)** -- Full docs and API reference
+- **[GitHub Repository](https://github.com/kreuzberg-dev/liter-llm)** -- Source, issues, and discussions
+- **[Provider Registry](https://github.com/kreuzberg-dev/liter-llm/blob/main/schemas/providers.json)** -- 142 supported providers
+
+Part of [kreuzberg.dev](https://kreuzberg.dev).
+
+## Contributing
+
+Contributions are welcome! See [CONTRIBUTING.md](https://github.com/kreuzberg-dev/liter-llm/blob/main/CONTRIBUTING.md) for guidelines.
+
+Join our [Discord community](https://discord.gg/xt9WY3GnKR) for questions and discussion.
+
+## License
+
+MIT -- see [LICENSE](https://github.com/kreuzberg-dev/liter-llm/blob/main/LICENSE) for details.

data/lib/liter_llm.rb

@@ -0,0 +1,8 @@
+# frozen_string_literal: true
+
+require 'liter_llm_rb'
+
+# LiterLlm is a Ruby binding for the Rust core library providing a unified
+# LLM client interface with streaming, tool calling, and provider routing.
+module LiterLlm
+end

data/sig/liter_llm.rbs

@@ -0,0 +1,416 @@
+# Type signatures for the LiterLlm Ruby binding.
+# Generated from crates/liter-llm-rb Rust source — keep in sync with the
+# Magnus extension and lib/liter_llm.rb public API.
+
+module LiterLlm
+  VERSION: String
+
+  # ─── Shared types ────────────────────────────────────────────────────────────
+
+  # Token usage counts returned in both chat and embedding responses.
+  type usage_response = { prompt_tokens: Integer, completion_tokens: Integer, total_tokens: Integer }
+
+  # ─── Content types ───────────────────────────────────────────────────────────
+
+  type image_url_param = { url: String, detail: ("low" | "high" | "auto")? }
+
+  # A single part of a multipart user message.
+  type content_part_param =
+    { type: "text", text: String } |
+    { type: "image_url", image_url: image_url_param }
+
+  # ─── Message types ───────────────────────────────────────────────────────────
+
+  # A single message in the conversation history.
+  # The `content` field is a plain string for system/tool/developer/function
+  # roles and either a string or a list of content parts for user messages.
+  type message_param = {
+    role: ("system" | "user" | "assistant" | "tool" | "developer" | "function"),
+    content: String | Array[content_part_param],
+    name: String?,
+    tool_call_id: String?
+  }
+
+  # ─── Tool / function call types ──────────────────────────────────────────────
+
+  type function_definition = {
+    name: String,
+    description: String?,
+    parameters: Hash[String, untyped]?,
+    strict: bool?
+  }
+
+  type tool_param = { type: "function", function: function_definition }
+
+  type specific_tool_choice = { type: "function", function: { name: String } }
+
+  type tool_choice_param = ("auto" | "required" | "none") | specific_tool_choice
+
+  type function_call = { name: String, arguments: String }
+
+  type tool_call = { id: String, type: "function", function: function_call }
+
+  # ─── Response format ─────────────────────────────────────────────────────────
+
+  type json_schema_format = {
+    name: String,
+    description: String?,
+    schema: Hash[String, untyped],
+    strict: bool?
+  }
+
+  type response_format_param =
+    { type: "text" } |
+    { type: "json_object" } |
+    { type: "json_schema", json_schema: json_schema_format }
+
+  # ─── Chat request / response ─────────────────────────────────────────────────
+
+  type stream_options = { include_usage: bool? }
+
+  # Full OpenAI-compatible chat completion request.
+  type chat_request = {
+    model: String,
+    messages: Array[message_param],
+    temperature: Float?,
+    top_p: Float?,
+    n: Integer?,
+    stream: bool?,
+    stop: (String | Array[String])?,
+    max_tokens: Integer?,
+    presence_penalty: Float?,
+    frequency_penalty: Float?,
+    logit_bias: Hash[String, Float]?,
+    user: String?,
+    tools: Array[tool_param]?,
+    tool_choice: tool_choice_param?,
+    parallel_tool_calls: bool?,
+    response_format: response_format_param?,
+    stream_options: stream_options?,
+    seed: Integer?
+  }
+
+  type assistant_message = {
+    content: String?,
+    name: String?,
+    tool_calls: Array[tool_call]?,
+    refusal: String?,
+    function_call: function_call?
+  }
+
+  type choice_response = {
+    index: Integer,
+    message: assistant_message,
+    finish_reason: ("stop" | "length" | "tool_calls" | "content_filter" | "function_call" | String)?
+  }
+
+  # Full OpenAI-compatible chat completion response.
+  type chat_response = {
+    id: String,
+    object: String,
+    created: Integer,
+    model: String,
+    choices: Array[choice_response],
+    usage: usage_response?,
+    system_fingerprint: String?,
+    service_tier: String?
+  }
+
+  # ─── Streaming chunk types ───────────────────────────────────────────────────
+
+  type stream_function_call = { name: String?, arguments: String? }
+
+  type stream_tool_call = {
+    index: Integer,
+    id: String?,
+    type: "function"?,
+    function: stream_function_call?
+  }
+
+  type stream_delta = {
+    role: String?,
+    content: String?,
+    tool_calls: Array[stream_tool_call]?,
+    function_call: stream_function_call?,
+    refusal: String?
+  }
+
+  type stream_choice = { index: Integer, delta: stream_delta, finish_reason: String? }
+
+  type chat_completion_chunk = {
+    id: String,
+    object: String,
+    created: Integer,
+    model: String,
+    choices: Array[stream_choice],
+    usage: usage_response?,
+    service_tier: String?
+  }
+
+  # ─── Embedding types ─────────────────────────────────────────────────────────
+
+  type embedding_request = {
+    model: String,
+    input: String | Array[String],
+    encoding_format: String?,
+    dimensions: Integer?,
+    user: String?
+  }
+
+  type embedding_object = { object: String, embedding: Array[Float], index: Integer }
+
+  type embedding_response = {
+    object: String,
+    data: Array[embedding_object],
+    model: String,
+    usage: usage_response
+  }
+
+  # ─── Models types ────────────────────────────────────────────────────────────
+
+  type model_object = { id: String, object: String, created: Integer, owned_by: String }
+
+  type models_response = { object: String, data: Array[model_object] }
+
+  # ─── Image Generation types ──────────────────────────────────────────────────
+
+  type create_image_request = {
+    prompt: String,
+    model: String?,
+    n: Integer?,
+    quality: String?,
+    response_format: String?,
+    size: String?,
+    style: String?,
+    user: String?
+  }
+
+  type image_data = { url: String?, b64_json: String?, revised_prompt: String? }
+
+  type images_response = { created: Integer, data: Array[image_data] }
+
+  # ─── Speech types ──────────────────────────────────────────────────────────
+
+  type create_speech_request = {
+    model: String,
+    input: String,
+    voice: String,
+    response_format: String?,
+    speed: Float?
+  }
+
+  # ─── Transcription types ───────────────────────────────────────────────────
+
+  type create_transcription_request = {
+    file: String,
+    model: String,
+    language: String?,
+    prompt: String?,
+    response_format: String?,
+    temperature: Float?
+  }
+
+  type transcription_response = { text: String }
+
+  # ─── Moderation types ──────────────────────────────────────────────────────
+
+  type moderation_request = {
+    input: untyped,
+    model: String?
+  }
+
+  type moderation_categories = {
+    sexual: bool,
+    hate: bool,
+    harassment: bool,
+    violence: bool
+  }
+
+  type moderation_category_scores = {
+    sexual: Float,
+    hate: Float,
+    harassment: Float,
+    violence: Float
+  }
+
+  type moderation_result = {
+    flagged: bool,
+    categories: moderation_categories,
+    category_scores: moderation_category_scores
+  }
+
+  type moderation_response = { id: String, model: String, results: Array[moderation_result] }
+
+  # ─── Rerank types ──────────────────────────────────────────────────────────
+
+  type rerank_request = {
+    model: String,
+    query: String,
+    documents: untyped,
+    top_n: Integer?
+  }
+
+  type rerank_result = { index: Integer, relevance_score: Float }
+
+  type rerank_response = { results: Array[rerank_result], model: String, usage: usage_response? }
+
+  # ─── File types ────────────────────────────────────────────────────────────
+
+  type create_file_request = { file: String, purpose: String, filename: String? }
+
+  type file_object = {
+    id: String,
+    object: String,
+    bytes: Integer,
+    created_at: Integer,
+    filename: String,
+    purpose: String,
+    status: String?,
+    status_details: String?
+  }
+
+  type delete_response = { id: String, object: String, deleted: bool }
+
+  type file_list_query = { purpose: String?, limit: Integer?, after: String? }
+
+  type file_list_response = { object: String, data: Array[file_object] }
+
+  # ─── Batch types ───────────────────────────────────────────────────────────
+
+  type create_batch_request = {
+    input_file_id: String,
+    endpoint: String,
+    completion_window: String,
+    metadata: Hash[String, String]?
+  }
+
+  type batch_request_counts = { total: Integer, completed: Integer, failed: Integer }
+
+  type batch_object = {
+    id: String,
+    object: String,
+    endpoint: String,
+    input_file_id: String,
+    completion_window: String,
+    status: String,
+    output_file_id: String?,
+    error_file_id: String?,
+    created_at: Integer,
+    request_counts: batch_request_counts?,
+    metadata: Hash[String, String]?
+  }
+
+  type batch_list_query = { limit: Integer?, after: String? }
+
+  type batch_list_response = { object: String, data: Array[batch_object] }
+
+  # ─── Response types ────────────────────────────────────────────────────────
+
+  type create_response_request = {
+    model: String,
+    input: untyped,
+    instructions: String?,
+    max_output_tokens: Integer?,
+    temperature: Float?,
+    top_p: Float?,
+    stream: bool?,
+    metadata: Hash[String, String]?
+  }
+
+  type response_object = {
+    id: String,
+    object: String,
+    created_at: Integer,
+    status: String,
+    model: String,
+    output: untyped?,
+    usage: usage_response?,
+    metadata: Hash[String, String]?,
+    error: untyped?
+  }
+
+  # ─── LlmClient ───────────────────────────────────────────────────────────────
+
+  # Unified LLM client backed by the Rust core.
+  #
+  # All I/O methods accept a JSON-encoded request string and return a
+  # JSON-encoded response string.  The thin Ruby layer is responsible for
+  # serialising/deserialising as needed.
+  class LlmClient
+    # Create a new client.
+    #
+    # @param api_key     API key for authentication.
+    # @param base_url    Optional provider base URL override.
+    # @param max_retries Retries on 429 / 5xx (default: 3).
+    # @param timeout_secs Request timeout in seconds (default: 60).
+    def initialize: (
+      String api_key,
+      ?base_url: String?,
+      ?model_hint: String?,
+      ?max_retries: Integer,
+      ?timeout_secs: Integer
+    ) -> void
+
+    # Send a chat completion request.
+    def chat: (String request_json) -> String
+
+    # Send an embedding request.
+    def embed: (String request_json) -> String
+
+    # List models available from the configured provider.
+    def list_models: () -> String
+
+    # Generate an image from a text prompt.
+    def image_generate: (String request_json) -> String
+
+    # Generate audio speech from text, returning base64-encoded audio bytes.
+    def speech: (String request_json) -> String
+
+    # Transcribe audio to text.
+    def transcribe: (String request_json) -> String
+
+    # Check content against moderation policies.
+    def moderate: (String request_json) -> String
+
+    # Rerank documents by relevance to a query.
+    def rerank: (String request_json) -> String
+
+    # Upload a file.
+    def create_file: (String request_json) -> String
+
+    # Retrieve metadata for a file by ID.
+    def retrieve_file: (String file_id) -> String
+
+    # Delete a file by ID.
+    def delete_file: (String file_id) -> String
+
+    # List files, optionally filtered by query parameters.
+    def list_files: (String? query_json) -> String
+
+    # Retrieve the raw content of a file (base64-encoded).
+    def file_content: (String file_id) -> String
+
+    # Create a new batch job.
+    def create_batch: (String request_json) -> String
+
+    # Retrieve a batch by ID.
+    def retrieve_batch: (String batch_id) -> String
+
+    # List batches, optionally filtered by query parameters.
+    def list_batches: (String? query_json) -> String
+
+    # Cancel an in-progress batch.
+    def cancel_batch: (String batch_id) -> String
+
+    # Create a new response via the Responses API.
+    def create_response: (String request_json) -> String
+
+    # Retrieve a response by ID.
+    def retrieve_response: (String response_id) -> String
+
+    # Cancel an in-progress response.
+    def cancel_response: (String response_id) -> String
+
+    def inspect: () -> String
+  end
+end

metadata

@@ -0,0 +1,199 @@
+--- !ruby/object:Gem::Specification
+name: liter_llm
+version: !ruby/object:Gem::Version
+  version: 1.0.0.pre.rc.6
+platform: aarch64-linux
+authors:
+- Na'aman Hirschfeld
+autorequire:
+bindir: bin
+cert_chain: []
+date: 2026-03-28 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '4.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '4.0'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '13.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '13.0'
+- !ruby/object:Gem::Dependency
+  name: rake-compiler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.2'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.2'
+- !ruby/object:Gem::Dependency
+  name: rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.12'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.12'
+- !ruby/object:Gem::Dependency
+  name: rbs
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+- !ruby/object:Gem::Dependency
+  name: rubocop
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.66'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.66'
+- !ruby/object:Gem::Dependency
+  name: rubocop-performance
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.21'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.21'
+- !ruby/object:Gem::Dependency
+  name: rubocop-rspec
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '3.0'
+- !ruby/object:Gem::Dependency
+  name: steep
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.8'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.8'
+- !ruby/object:Gem::Dependency
+  name: yard
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.9'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.9'
+description: |
+  liter-llm is a universal LLM API client with a Rust core and native Ruby bindings
+  via Magnus. Provides a unified interface for streaming completions, tool calling,
+  and provider routing across 142+ LLM providers. Rust-powered.
+email:
+- naaman@kreuzberg.dev
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- README.md
+- lib/liter_llm.rb
+- lib/liter_llm_rb.so
+- sig/liter_llm.rbs
+homepage: https://kreuzberg.dev
+licenses:
+- MIT
+metadata:
+  homepage_uri: https://kreuzberg.dev
+  source_code_uri: https://github.com/kreuzberg-dev/liter-llm
+  changelog_uri: https://github.com/kreuzberg-dev/liter-llm/blob/main/CHANGELOG.md
+  bug_tracker_uri: https://github.com/kreuzberg-dev/liter-llm/issues
+  rubygems_mfa_required: 'true'
+  keywords: llm,llm-client,openai,anthropic,streaming,tool-calling,provider-routing,rust,native-extension
+post_install_message:
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: 3.2.0
+  - - "<"
+    - !ruby/object:Gem::Version
+      version: '5.0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.5.22
+signing_key:
+specification_version: 4
+summary: Universal LLM API client — 142+ providers, streaming, tool calling. Rust-powered.
+test_files: []