llm_ruby 0.1.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: dfe908817dd406ae16aca4130133b9a421b18333cdecc4ad870635dd997be500
+  data.tar.gz: 105ae0dcc30918686abcf8d01d99605d2f70f41eebdd2737744bc2bf27c6575c
+SHA512:
+  metadata.gz: 5b9643df8771735111f18f52182b6f217231ca92c890e3456f21c3937850bf2c9ac668730cd1aeae8b43330d5a3c84eab6b419c60aa5d54d70f405793e4463ad
+  data.tar.gz: 820687838675aeaadde8e5b7c5b7f7f45bfdf10beb74b90e2b594cd80d5e7accc38ace2deec56a6a2fcdc3de6a369af41405c5523da3e4c2b47f6e584c28f3fd

data/.rspec

@@ -0,0 +1,3 @@
+--format documentation
+--color
+--require spec_helper

data/.standard.yml

@@ -0,0 +1,3 @@
+# For available configuration options, see:
+#   https://github.com/standardrb/standard
+ruby_version: 3.1.0

data/LICENSE.txt

@@ -0,0 +1,21 @@
+The MIT License (MIT)
+
+Copyright (c) 2024 TODO: Write your name
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.

data/README.md

@@ -0,0 +1,153 @@
+# LLMRuby
+
+LLMRuby is a Ruby gem that provides a consistent interface for interacting with various Large Language Model (LLM) APIs, with a current focus on OpenAI's models.
+
+## Installation
+
+Add this line to your application's Gemfile:
+
+```ruby
+gem 'llm_ruby'
+```
+
+And then execute:
+
+```
+$ bundle install
+```
+
+Or install it yourself as:
+
+```
+$ gem install llm_ruby
+```
+
+## Usage
+
+### Basic Usage
+
+```ruby
+require 'llm'
+
+# Initialize an LLM instance
+llm = LLM.from_string!("gpt-4")
+
+# Create a client
+client = llm.client
+
+# Send a chat message
+response = client.chat([{role: :user, content: "Hello, world!"}])
+
+puts response.content
+```
+
+### Streaming Responses
+
+LLMRuby supports streaming responses:
+
+```ruby
+require 'llm'
+
+# Initialize an LLM instance
+llm = LLM.from_string!("gpt-4")
+
+# Create a client
+client = llm.client
+
+# Define the on_message callback
+on_message = proc do |message|
+  puts "Received message chunk: #{message}"
+end
+
+# Define the on_complete callback
+on_complete = proc do |stop_reason|
+  puts "Streaming complete. Stop reason: #{stop_reason}"
+end
+
+# Send a chat message with streaming enabled
+response = client.chat(
+  [{role: :user, content: "Hello, world!"}],
+  stream: true,
+  on_message: on_message,
+  on_complete: on_complete
+)
+
+puts response.content
+```
+
+### Using the Response Object
+
+The response object returned by the `client.chat` method contains several useful fields:
+
+- `content`: The final content of the response.
+- `raw_response`: The raw response payload for non-streaming requests and the array of chunks for streaming requests.
+- `stop_reason`: The reason why the response generation was stopped.
+
+Here is an example of how to use the response object:
+
+```ruby
+# Initialize an LLM instance
+llm = LLM.from_string!("gpt-4")
+
+# Create a client
+client = llm.client
+
+# Send a chat message
+response = client.chat([{role: :user, content: "Hello, world!"}])
+
+# Access the response fields
+puts "Response content: #{response.content}"
+puts "Raw response: #{response.raw_response}"
+puts "Stop reason: #{response.stop_reason}"
+```
+
+
+## Available Models
+
+LLMRuby supports various OpenAI models, including GPT-3.5 and GPT-4 variants. You can see the full list of supported models in the `KNOWN_MODELS` constant:
+
+| Canonical Name            | Display Name           | Provider |
+|---------------------------|------------------------|----------|
+| gpt-3.5-turbo             | GPT-3.5 Turbo          | openai   |
+| gpt-3.5-turbo-0125        | GPT-3.5 Turbo 0125     | openai   |
+| gpt-3.5-turbo-16k         | GPT-3.5 Turbo 16K      | openai   |
+| gpt-3.5-turbo-1106        | GPT-3.5 Turbo 1106     | openai   |
+| gpt-4                     | GPT-4                  | openai   |
+| gpt-4-32k                 | GPT-4 32K              | openai   |
+| gpt-4-1106-preview        | GPT-4 Turbo 1106       | openai   |
+| gpt-4-turbo-2024-04-09    | GPT-4 Turbo 2024-04-09 | openai   |
+| gpt-4-0125-preview        | GPT-4 Turbo 0125       | openai   |
+| gpt-4-turbo-preview       | GPT-4 Turbo            | openai   |
+| gpt-4-0613                | GPT-4 0613             | openai   |
+| gpt-4-32k-0613            | GPT-4 32K 0613         | openai   |
+| gpt-4o                    | GPT-4o                 | openai   |
+| gpt-4o-mini               | GPT-4o Mini            | openai   |
+| gpt-4o-2024-05-13         | GPT-4o 2024-05-13      | openai   |
+| gpt-4o-2024-08-06         | GPT-4o 2024-08-06      | openai   |
+
+
+## Configuration
+
+Set your OpenAI API key as an environment variable:
+
+```
+export OPENAI_API_KEY=your_api_key_here
+```
+
+## Development
+
+After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake spec` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.
+
+To install this gem onto your local machine, run `bundle exec rake install`.
+
+## Contributing
+
+Bug reports and pull requests are welcome on GitHub at https://github.com/contextco/llm_ruby.
+
+## License
+
+The gem is available as open source under the terms of the [MIT License](https://opensource.org/licenses/MIT).
+
+## Acknowledgements
+
+This gem is developed and maintained by [Context](https://context.ai).

data/Rakefile

@@ -0,0 +1,10 @@
+# frozen_string_literal: true
+
+require "bundler/gem_tasks"
+require "rspec/core/rake_task"
+
+RSpec::Core::RakeTask.new(:spec)
+
+require "standard/rake"
+
+task default: %i[spec standard]

data/lib/llm/clients/open_ai/response.rb

@@ -0,0 +1,42 @@
+# frozen_string_literal: true
+
+class LLM::Clients::OpenAI::Response
+  def initialize(raw_response)
+    @raw_response = raw_response
+  end
+
+  def to_normalized_response
+    LLM::Response.new(
+      content: content,
+      raw_response: parsed_response,
+      stop_reason: normalize_stop_reason
+    )
+  end
+
+  def self.normalize_stop_reason(stop_reason)
+    case stop_reason
+    when "stop"
+      LLM::StopReason::STOP
+    when "safety"
+      LLM::StopReason::SAFETY
+    when "max_tokens"
+      LLM::StopReason::MAX_TOKENS_REACHED
+    else
+      LLM::StopReason::OTHER
+    end
+  end
+
+  private
+
+  def content
+    parsed_response.dig("choices", 0, "message", "content")
+  end
+
+  def normalize_stop_reason
+    self.class.normalize_stop_reason(parsed_response.dig("choices", 0, "finish_reason"))
+  end
+
+  def parsed_response
+    @raw_response.parsed_response
+  end
+end

data/lib/llm/clients/open_ai.rb

@@ -0,0 +1,109 @@
+# frozen_string_literal: true
+
+require "httparty"
+require "event_stream_parser"
+
+class LLM::Clients::OpenAI
+  include HTTParty
+  base_uri "https://api.openai.com/v1"
+
+  def initialize(llm:)
+    @llm = llm
+  end
+
+  def chat(messages, options = {})
+    parameters = {
+      model: @llm.canonical_name,
+      messages: messages,
+      temperature: options[:temperature],
+      response_format: options[:response_format],
+      max_tokens: options[:max_output_tokens],
+      top_p: options[:top_p],
+      stop: options[:stop_sequences],
+      presence_penalty: options[:presence_penalty],
+      frequency_penalty: options[:frequency_penalty],
+      tools: options[:tools],
+      tool_choice: options[:tool_choice]
+    }.compact
+
+    return chat_streaming(parameters, options[:on_message], options[:on_complete]) if options[:stream]
+
+    resp = post_url("/chat/completions", body: parameters.to_json)
+
+    Response.new(resp).to_normalized_response
+  end
+
+  private
+
+  def chat_streaming(parameters, on_message, on_complete)
+    buffer = +""
+    chunks = []
+    output_data = {}
+
+    wrapped_on_complete = lambda { |stop_reason|
+      output_data[:stop_reason] = stop_reason
+      on_complete&.call(stop_reason)
+    }
+
+    parameters[:stream] = true
+
+    proc = stream_proc(buffer, chunks, on_message, wrapped_on_complete)
+
+    parameters.delete(:on_message)
+    parameters.delete(:on_complete)
+
+    _resp = post_url_streaming("/chat/completions", body: parameters.to_json, &proc)
+
+    LLM::Response.new(
+      content: buffer,
+      raw_response: chunks,
+      stop_reason: output_data[:stop_reason]
+    )
+  end
+
+  def stream_proc(buffer, chunks, on_message, complete_proc)
+    each_json_chunk do |_type, event|
+      next if event == "[DONE]"
+
+      chunks << event
+      new_content = event.dig("choices", 0, "delta", "content")
+      stop_reason = event.dig("choices", 0, "finish_reason")
+
+      buffer << new_content unless new_content.nil?
+      on_message&.call(new_content) unless new_content.nil?
+      complete_proc&.call(Response.normalize_stop_reason(stop_reason)) unless stop_reason.nil?
+    end
+  end
+
+  def each_json_chunk
+    parser = EventStreamParser::Parser.new
+
+    proc do |chunk, _bytes, env|
+      if env && env.status != 200
+        raise_error = Faraday::Response::RaiseError.new
+        raise_error.on_complete(env.merge(body: try_parse_json(chunk)))
+      end
+
+      parser.feed(chunk) do |type, data|
+        next if data == "[DONE]"
+
+        yield(type, JSON.parse(data))
+      end
+    end
+  end
+
+  def post_url(url, **kwargs)
+    self.class.post(url, **kwargs.merge(headers: default_headers))
+  end
+
+  def post_url_streaming(url, **kwargs, &block)
+    self.class.post(url, **kwargs.merge(headers: default_headers, stream_body: true), &block)
+  end
+
+  def default_headers
+    {
+      "Authorization" => "Bearer #{ENV["OPENAI_API_KEY"]}",
+      "Content-Type" => "application/json"
+    }
+  end
+end

data/lib/llm/info.rb

@@ -0,0 +1,94 @@
+# frozen_string_literal: true
+
+module LLM::Info
+  KNOWN_MODELS = [
+    # Semantics of fields:
+    # - canonical_name (required): A string that uniquely identifies the model.
+    #   We use this string as the public identifier when users choose this model via the API.
+    # - display_name (required): A string that is displayed to the user when choosing this model via the UI.
+
+    # GPT-3.5 Turbo Models
+    {
+      canonical_name: "gpt-3.5-turbo",
+      display_name: "GPT-3.5 Turbo",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-3.5-turbo-0125",
+      display_name: "GPT-3.5 Turbo 0125",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-3.5-turbo-16k",
+      display_name: "GPT-3.5 Turbo 16K",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-3.5-turbo-1106",
+      display_name: "GPT-3.5 Turbo 1106",
+      provider: :openai
+    },
+
+    # GPT-4 Models
+    {
+      canonical_name: "gpt-4",
+      display_name: "GPT-4",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4-32k",
+      display_name: "GPT-4 32K",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4-1106-preview",
+      display_name: "GPT-4 Turbo 1106",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4-turbo-2024-04-09",
+      display_name: "GPT-4 Turbo 2024-04-09",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4-0125-preview",
+      display_name: "GPT-4 Turbo 0125",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4-turbo-preview",
+      display_name: "GPT-4 Turbo",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4-0613",
+      display_name: "GPT-4 0613",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4-32k-0613",
+      display_name: "GPT-4 32K 0613",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4o",
+      display_name: "GPT-4o",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4o-mini",
+      display_name: "GPT-4o Mini",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4o-2024-05-13",
+      display_name: "GPT-4o 2024-05-13",
+      provider: :openai
+    },
+    {
+      canonical_name: "gpt-4o-2024-08-06",
+      display_name: "GPT-4o 2024-08-06",
+      provider: :openai
+    }
+  ].freeze
+end

data/lib/llm/response.rb

@@ -0,0 +1,3 @@
+# frozen_string_literal: true
+
+LLM::Response = Struct.new(:content, :raw_response, :stop_reason, keyword_init: true)

data/lib/llm/stop_reason.rb

@@ -0,0 +1,9 @@
+# frozen_string_literal: true
+
+module LLM::StopReason
+  STOP = :stop
+  SAFETY = :safety
+  MAX_TOKENS_REACHED = :max_tokens
+
+  OTHER = :other
+end

data/lib/llm/version.rb

@@ -0,0 +1 @@
+# frozen_string_literal: true

data/lib/llm.rb

@@ -0,0 +1,50 @@
+# frozen_string_literal: true
+
+require "zeitwerk"
+loader = Zeitwerk::Loader.for_gem
+loader.inflector.inflect(
+  "llm" => "LLM",
+  "open_ai" => "OpenAI"
+)
+loader.setup
+
+class LLM
+  def initialize(model)
+    @canonical_name = model[:canonical_name]
+    @display_name = model[:display_name]
+    @provider = model[:provider]
+    @client_class = LLM::Clients::OpenAI # TODO: Allow alternative client classes.
+  end
+
+  def client
+    client_class.new(llm: self)
+  end
+
+  attr_reader :canonical_name,
+    :display_name,
+    :provider
+
+  private
+
+  attr_reader :client_class
+
+  class << self
+    def all!
+      known_models
+    end
+
+    def from_string!(model_string)
+      model = known_models.find { |model| model.canonical_name == model_string }
+
+      raise ArgumentError, "Unknown model: #{model_string}" unless model
+
+      model
+    end
+
+    private
+
+    def known_models
+      @known_models ||= LLM::Info::KNOWN_MODELS.map { |model| new(model) }
+    end
+  end
+end

metadata

@@ -0,0 +1,185 @@
+--- !ruby/object:Gem::Specification
+name: llm_ruby
+version: !ruby/object:Gem::Version
+  version: 0.1.0
+platform: ruby
+authors:
+- Alex Gamble
+autorequire: 
+bindir: exe
+cert_chain: []
+date: 2024-09-13 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: event_stream_parser
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.0.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.0.0
+- !ruby/object:Gem::Dependency
+  name: httparty
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 0.22.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 0.22.0
+- !ruby/object:Gem::Dependency
+  name: zeitwerk
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 2.6.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 2.6.0
+- !ruby/object:Gem::Dependency
+  name: dotenv
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 3.1.0
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 3.1.0
+- !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: 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: standard
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.31.0
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 1.31.0
+- !ruby/object:Gem::Dependency
+  name: vcr
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 6.3.1
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 6.3.1
+- !ruby/object:Gem::Dependency
+  name: webmock
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 3.23.0
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: 3.23.0
+description: 
+email:
+- alex@context.ai
+- alec@context.ai
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- ".rspec"
+- ".standard.yml"
+- LICENSE.txt
+- README.md
+- Rakefile
+- lib/llm.rb
+- lib/llm/clients/open_ai.rb
+- lib/llm/clients/open_ai/response.rb
+- lib/llm/info.rb
+- lib/llm/provider.rb
+- lib/llm/response.rb
+- lib/llm/stop_reason.rb
+- lib/llm/version.rb
+homepage: https://context.ai
+licenses:
+- MIT
+metadata:
+  homepage_uri: https://context.ai
+  source_code_uri: https://github.com/contextco/llm_ruby
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: 3.1.0
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.5.16
+signing_key: 
+specification_version: 4
+summary: A client to interact with LLM APIs in a consistent way.
+test_files: []