ruby_llm 1.6.3 → 1.7.0

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: ruby_llm
 version: !ruby/object:Gem::Version
-  version: 1.6.3
+  version: 1.7.0
 platform: ruby
 authors:
 - Carmine Paolino
@@ -136,17 +136,40 @@ extra_rdoc_files: []
 files:
 - LICENSE
 - README.md
-- lib/generators/ruby_llm/install/templates/INSTALL_INFO.md.tt
+- lib/generators/ruby_llm/chat_ui/chat_ui_generator.rb
+- lib/generators/ruby_llm/chat_ui/templates/controllers/chats_controller.rb.tt
+- lib/generators/ruby_llm/chat_ui/templates/controllers/messages_controller.rb.tt
+- lib/generators/ruby_llm/chat_ui/templates/controllers/models_controller.rb.tt
+- lib/generators/ruby_llm/chat_ui/templates/jobs/chat_response_job.rb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/chats/_chat.html.erb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/chats/_form.html.erb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/chats/index.html.erb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/chats/new.html.erb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/chats/show.html.erb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/messages/_form.html.erb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/messages/_message.html.erb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/messages/create.turbo_stream.erb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/models/_model.html.erb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/models/index.html.erb.tt
+- lib/generators/ruby_llm/chat_ui/templates/views/models/show.html.erb.tt
 - lib/generators/ruby_llm/install/templates/chat_model.rb.tt
 - lib/generators/ruby_llm/install/templates/create_chats_migration.rb.tt
 - lib/generators/ruby_llm/install/templates/create_messages_migration.rb.tt
+- lib/generators/ruby_llm/install/templates/create_models_migration.rb.tt
 - lib/generators/ruby_llm/install/templates/create_tool_calls_migration.rb.tt
 - lib/generators/ruby_llm/install/templates/initializer.rb.tt
 - lib/generators/ruby_llm/install/templates/message_model.rb.tt
+- lib/generators/ruby_llm/install/templates/model_model.rb.tt
 - lib/generators/ruby_llm/install/templates/tool_call_model.rb.tt
 - lib/generators/ruby_llm/install_generator.rb
+- lib/generators/ruby_llm/upgrade_to_v1_7/templates/migration.rb.tt
+- lib/generators/ruby_llm/upgrade_to_v1_7_generator.rb
 - lib/ruby_llm.rb
 - lib/ruby_llm/active_record/acts_as.rb
+- lib/ruby_llm/active_record/acts_as_legacy.rb
+- lib/ruby_llm/active_record/chat_methods.rb
+- lib/ruby_llm/active_record/message_methods.rb
+- lib/ruby_llm/active_record/model_methods.rb
 - lib/ruby_llm/aliases.json
 - lib/ruby_llm/aliases.rb
 - lib/ruby_llm/attachment.rb
@@ -205,6 +228,7 @@ files:
 - lib/ruby_llm/providers/gemini/tools.rb
 - lib/ruby_llm/providers/gpustack.rb
 - lib/ruby_llm/providers/gpustack/chat.rb
+- lib/ruby_llm/providers/gpustack/media.rb
 - lib/ruby_llm/providers/gpustack/models.rb
 - lib/ruby_llm/providers/mistral.rb
 - lib/ruby_llm/providers/mistral/capabilities.rb
@@ -214,6 +238,7 @@ files:
 - lib/ruby_llm/providers/ollama.rb
 - lib/ruby_llm/providers/ollama/chat.rb
 - lib/ruby_llm/providers/ollama/media.rb
+- lib/ruby_llm/providers/ollama/models.rb
 - lib/ruby_llm/providers/openai.rb
 - lib/ruby_llm/providers/openai/capabilities.rb
 - lib/ruby_llm/providers/openai/chat.rb
@@ -229,6 +254,11 @@ files:
 - lib/ruby_llm/providers/perplexity/capabilities.rb
 - lib/ruby_llm/providers/perplexity/chat.rb
 - lib/ruby_llm/providers/perplexity/models.rb
+- lib/ruby_llm/providers/vertexai.rb
+- lib/ruby_llm/providers/vertexai/chat.rb
+- lib/ruby_llm/providers/vertexai/embeddings.rb
+- lib/ruby_llm/providers/vertexai/models.rb
+- lib/ruby_llm/providers/vertexai/streaming.rb
 - lib/ruby_llm/railtie.rb
 - lib/ruby_llm/stream_accumulator.rb
 - lib/ruby_llm/streaming.rb
@@ -236,10 +266,9 @@ files:
 - lib/ruby_llm/tool_call.rb
 - lib/ruby_llm/utils.rb
 - lib/ruby_llm/version.rb
-- lib/tasks/aliases.rake
-- lib/tasks/models_docs.rake
-- lib/tasks/models_update.rake
+- lib/tasks/models.rake
 - lib/tasks/release.rake
+- lib/tasks/ruby_llm.rake
 - lib/tasks/vcr.rake
 homepage: https://rubyllm.com
 licenses:
@@ -251,6 +280,9 @@ metadata:
   documentation_uri: https://rubyllm.com
   bug_tracker_uri: https://github.com/crmne/ruby_llm/issues
   rubygems_mfa_required: 'true'
+post_install_message: |
+  Upgrading from RubyLLM <= 1.6.x? Check the upgrade guide for new features and migration instructions
+  --> https://rubyllm.com/upgrading-to-1-7/
 rdoc_options: []
 require_paths:
 - lib

data/lib/generators/ruby_llm/install/templates/INSTALL_INFO.md.tt

@@ -1,108 +0,0 @@
-# RubyLLM Rails Setup Complete!
-
-Thanks for installing RubyLLM in your Rails application. Here's what was created:
-
-## Models
-
-- `<%= options[:chat_model_name] %>` - Stores chat sessions and their associated model ID
-- `<%= options[:message_model_name] %>` - Stores individual messages in a chat
-- `<%= options[:tool_call_model_name] %>` - Stores tool calls made by language models
-
-**Note:** Do not add `validates :content, presence: true` to your Message model - RubyLLM creates empty assistant messages before API calls for streaming support.
-
-## Configuration Options
-
-The generator supports the following options to customize model names:
-
-```bash
-rails generate ruby_llm:install \
-  --chat-model-name=Conversation \
-  --message-model-name=ChatMessage \
-  --tool-call-model-name=FunctionCall
-```
-
-This is useful when you need to avoid namespace collisions with existing models in your application. Table names will be automatically derived from the model names following Rails conventions.
-
-## Next Steps
-
-1. **Run migrations:**
-   ```bash
-   rails db:migrate
-   ```
-
-   **Database Note:** The migrations use `jsonb` for PostgreSQL and `json` for MySQL/SQLite automatically.
-
-2. **Set your API keys** in `config/initializers/ruby_llm.rb` or using environment variables:
-   ```ruby
-   # config/initializers/ruby_llm.rb
-   RubyLLM.configure do |config|
-     config.openai_api_key = ENV['OPENAI_API_KEY']
-     config.anthropic_api_key = ENV['ANTHROPIC_API_KEY']
-     config.gemini_api_key = ENV['GEMINI_API_KEY']
-     # ... add other providers as needed
-   end
-   ```
-
-3. **Start using RubyLLM in your code:**
-   ```ruby
-   # Basic usage
-   chat = <%= options[:chat_model_name] %>.create!(model_id: 'gpt-4.1-nano')
-   response = chat.ask("What is Ruby on Rails?")
-
-   # With file attachments (requires ActiveStorage setup)
-   chat.ask("What's in this file?", with: "report.pdf")
-   chat.ask("Analyze these files", with: ["image.jpg", "data.csv", "notes.txt"])
-   ```
-
-4. **For streaming responses** with Hotwire/Turbo:
-   ```ruby
-   # app/models/<%= options[:message_model_name].underscore %>.rb
-   class <%= options[:message_model_name] %> < ApplicationRecord
-     acts_as_message
-
-     # Helper to broadcast chunks during streaming
-     def broadcast_append_chunk(chunk_content)
-       broadcast_append_to [ chat, "messages" ],
-         target: dom_id(self, "content"),
-         html: chunk_content
-     end
-   end
-
-   # app/jobs/chat_stream_job.rb
-   class ChatStreamJob < ApplicationJob
-     def perform(chat_id, user_content)
-       chat = <%= options[:chat_model_name] %>.find(chat_id)
-       chat.ask(user_content) do |chunk|
-         assistant_message = chat.messages.last
-         if chunk.content && assistant_message
-           assistant_message.broadcast_append_chunk(chunk.content)
-         end
-       end
-     end
-   end
-
-   # In your controller
-   ChatStreamJob.perform_later(@chat.id, params[:content])
-   ```
-
-## Optional: ActiveStorage for Attachments
-
-If you want to use file attachments (PDFs, images, etc.), set up ActiveStorage:
-
-```bash
-rails active_storage:install
-rails db:migrate
-```
-
-Then add to your Message model:
-```ruby
-class <%= options[:message_model_name] %> < ApplicationRecord
-  acts_as_message
-  has_many_attached :attachments
-end
-```
-
-## Learn More
-
-- See the [Rails Integration Guide](https://rubyllm.com/guides/rails) for detailed examples
-- Visit the [RubyLLM Documentation](https://rubyllm.com) for full API reference

data/lib/tasks/aliases.rake

@@ -1,205 +0,0 @@
-# frozen_string_literal: true
-
-require 'json'
-
-namespace :aliases do # rubocop:disable Metrics/BlockLength
-  desc 'Generate aliases.json from models in the registry'
-  task :generate do # rubocop:disable Metrics/BlockLength
-    require 'ruby_llm'
-
-    # Group models by provider
-    models = Hash.new { |h, k| h[k] = [] }
-
-    RubyLLM.models.all.each do |model|
-      models[model.provider] << model.id
-    end
-
-    aliases = {}
-
-    # OpenAI models
-    models['openai'].each do |model|
-      openrouter_model = "openai/#{model}"
-      next unless models['openrouter'].include?(openrouter_model)
-
-      alias_key = model.gsub('-latest', '')
-      aliases[alias_key] = {
-        'openai' => model,
-        'openrouter' => openrouter_model
-      }
-    end
-
-    anthropic_latest = group_anthropic_models_by_base_name(models['anthropic'])
-
-    anthropic_latest.each do |base_name, latest_model|
-      openrouter_variants = [
-        "anthropic/#{base_name}", # anthropic/claude-3-5-sonnet
-        "anthropic/#{base_name.gsub(/-(\d)/, '.\1')}", # anthropic/claude-3.5-sonnet
-        "anthropic/#{base_name.gsub(/claude-(\d+)-(\d+)/, 'claude-\1.\2')}", # claude-3-5 -> claude-3.5
-        "anthropic/#{base_name.gsub(/(\d+)-(\d+)/, '\1.\2')}" # any X-Y -> X.Y pattern
-      ]
-
-      openrouter_model = openrouter_variants.find { |v| models['openrouter'].include?(v) }
-
-      bedrock_model = find_best_bedrock_model(latest_model, models['bedrock'])
-
-      next unless openrouter_model || bedrock_model || models['anthropic'].include?(latest_model)
-
-      aliases[base_name] = {
-        'anthropic' => latest_model
-      }
-
-      aliases[base_name]['openrouter'] = openrouter_model if openrouter_model
-      aliases[base_name]['bedrock'] = bedrock_model if bedrock_model
-    end
-
-    models['bedrock'].each do |bedrock_model|
-      next unless bedrock_model.start_with?('anthropic.')
-
-      next unless bedrock_model =~ /anthropic\.(claude-[\d\.]+-[a-z]+)/
-
-      base_name = Regexp.last_match(1)
-      anthropic_name = base_name.tr('.', '-')
-
-      next if aliases[anthropic_name]
-
-      openrouter_variants = [
-        "anthropic/#{anthropic_name}",
-        "anthropic/#{base_name}" # Keep the dots
-      ]
-
-      openrouter_model = openrouter_variants.find { |v| models['openrouter'].include?(v) }
-
-      aliases[anthropic_name] = {
-        'bedrock' => bedrock_model
-      }
-
-      aliases[anthropic_name]['anthropic'] = anthropic_name if models['anthropic'].include?(anthropic_name)
-      aliases[anthropic_name]['openrouter'] = openrouter_model if openrouter_model
-    end
-
-    models['gemini'].each do |model|
-      openrouter_variants = [
-        "google/#{model}",
-        "google/#{model.gsub('gemini-', 'gemini-').tr('.', '-')}",
-        "google/#{model.gsub('gemini-', 'gemini-')}"
-      ]
-
-      openrouter_model = openrouter_variants.find { |v| models['openrouter'].include?(v) }
-      next unless openrouter_model
-
-      alias_key = model.gsub('-latest', '')
-      aliases[alias_key] = {
-        'gemini' => model,
-        'openrouter' => openrouter_model
-      }
-    end
-
-    models['deepseek'].each do |model|
-      openrouter_model = "deepseek/#{model}"
-      next unless models['openrouter'].include?(openrouter_model)
-
-      alias_key = model.gsub('-latest', '')
-      aliases[alias_key] = {
-        'deepseek' => model,
-        'openrouter' => openrouter_model
-      }
-    end
-
-    sorted_aliases = aliases.sort.to_h
-    File.write(RubyLLM::Aliases.aliases_file, JSON.pretty_generate(sorted_aliases))
-
-    puts "Generated #{sorted_aliases.size} aliases"
-  end
-
-  def group_anthropic_models_by_base_name(anthropic_models) # rubocop:disable Rake/MethodDefinitionInTask
-    grouped = Hash.new { |h, k| h[k] = [] }
-
-    anthropic_models.each do |model|
-      base_name = extract_base_name(model)
-      grouped[base_name] << model
-    end
-
-    latest_models = {}
-    grouped.each do |base_name, model_list|
-      if model_list.size == 1
-        latest_models[base_name] = model_list.first
-      else
-        latest_model = model_list.max_by { |model| extract_date_from_model(model) }
-        latest_models[base_name] = latest_model
-      end
-    end
-
-    latest_models
-  end
-
-  def extract_base_name(model) # rubocop:disable Rake/MethodDefinitionInTask
-    if model =~ /^(.+)-(\d{8})$/
-      Regexp.last_match(1)
-    else
-      model
-    end
-  end
-
-  def extract_date_from_model(model) # rubocop:disable Rake/MethodDefinitionInTask
-    if model =~ /-(\d{8})$/
-      Regexp.last_match(1)
-    else
-      '00000000'
-    end
-  end
-
-  def find_best_bedrock_model(anthropic_model, bedrock_models) # rubocop:disable Metrics/PerceivedComplexity,Rake/MethodDefinitionInTask
-    base_pattern = case anthropic_model
-                   when 'claude-2.0', 'claude-2'
-                     'claude-v2'
-                   when 'claude-2.1'
-                     'claude-v2:1'
-                   when 'claude-instant-v1', 'claude-instant'
-                     'claude-instant'
-                   else
-                     extract_base_name(anthropic_model)
-                   end
-
-    matching_models = bedrock_models.select do |bedrock_model|
-      model_without_prefix = bedrock_model.sub(/^(?:us\.)?anthropic\./, '')
-      model_without_prefix.start_with?(base_pattern)
-    end
-
-    return nil if matching_models.empty?
-
-    begin
-      model_info = RubyLLM.models.find(anthropic_model)
-      target_context = model_info.context_window
-    rescue StandardError
-      target_context = nil
-    end
-
-    if target_context
-      target_k = target_context / 1000
-
-      with_context = matching_models.select do |m|
-        m.include?(":#{target_k}k") || m.include?(":0:#{target_k}k")
-      end
-
-      return with_context.first if with_context.any?
-    end
-
-    matching_models.min_by do |model|
-      context_priority = if model =~ /:(?:\d+:)?(\d+)k/
-                           -Regexp.last_match(1).to_i
-                         else
-                           0
-                         end
-
-      version_priority = if model =~ /-v(\d+):/
-                           -Regexp.last_match(1).to_i
-                         else
-                           0
-                         end
-
-      # Prefer models with explicit context windows
-      has_context_priority = model.include?('k') ? -1 : 0
-      [has_context_priority, context_priority, version_priority]
-    end
-  end
-end

data/lib/tasks/models_docs.rake

@@ -1,214 +0,0 @@
-# frozen_string_literal: true
-
-require 'dotenv/load'
-require 'fileutils'
-
-namespace :models do
-  desc 'Generate available models documentation'
-  task :docs do
-    FileUtils.mkdir_p('docs')
-    output = generate_models_markdown
-    File.write('docs/_reference/available-models.md', output)
-    puts 'Generated docs/_reference/available-models.md'
-  end
-end
-
-def generate_models_markdown
-  <<~MARKDOWN
-    ---
-    layout: default
-    title: Available Models
-    nav_order: 1
-    description: Browse hundreds of AI models from every major provider. Always up-to-date, automatically generated.
-    redirect_from:
-      - /guides/available-models
-    ---
-
-    # {{ page.title }}
-    {: .no_toc }
-
-    {{ page.description }}
-    {: .fs-6 .fw-300 }
-
-    ## Table of contents
-    {: .no_toc .text-delta }
-
-    1. TOC
-    {:toc}
-
-    ---
-
-    After reading this guide, you will know:
-
-    * How RubyLLM's model registry works and where data comes from
-    * How to find models by provider, capability, or purpose
-    * What information is available for each model
-    * How to use model aliases for simpler configuration
-
-    ## How Model Data Works
-
-    RubyLLM's model registry combines data from multiple sources:
-
-    - **OpenAI, Anthropic, DeepSeek, Gemini**: Data from [Parsera](https://api.parsera.org/v1/llm-specs)
-    - **OpenRouter**: Direct from OpenRouter's API
-    - **Other providers**: Defined in `capabilities.rb` files
-
-    ## Contributing Model Updates
-
-    **For major providers** (OpenAI, Anthropic, DeepSeek, Gemini): File issues with [Parsera](https://github.com/parsera-labs/api-llm-specs/issues) for public model data corrections.
-
-    **For other providers**: Edit `lib/ruby_llm/providers/<provider>/capabilities.rb` then run `rake models:update`.
-
-    See the [Contributing Guide](https://github.com/crmne/ruby_llm/blob/main/CONTRIBUTING.md) for details.
-
-    ## Last Updated
-    {: .d-inline-block }
-
-    #{Time.now.utc.strftime('%Y-%m-%d')}
-    {: .label .label-green }
-
-    ## Models by Provider
-
-    #{generate_provider_sections}
-
-    ## Models by Capability
-
-    #{generate_capability_sections}
-
-    ## Models by Modality
-
-    #{generate_modality_sections}
-  MARKDOWN
-end
-
-def generate_provider_sections
-  RubyLLM::Provider.providers.filter_map do |provider, provider_class|
-    models = RubyLLM.models.by_provider(provider)
-    next if models.none?
-
-    <<~PROVIDER
-      ### #{provider_class.name} (#{models.count})
-
-      #{models_table(models)}
-    PROVIDER
-  end.join("\n\n")
-end
-
-def generate_capability_sections
-  capabilities = {
-    'Function Calling' => RubyLLM.models.select(&:function_calling?),
-    'Structured Output' => RubyLLM.models.select(&:structured_output?),
-    'Streaming' => RubyLLM.models.select { |m| m.capabilities.include?('streaming') },
-    # 'Reasoning' => RubyLLM.models.select { |m| m.capabilities.include?('reasoning') },
-    'Batch Processing' => RubyLLM.models.select { |m| m.capabilities.include?('batch') }
-  }
-
-  capabilities.filter_map do |capability, models|
-    next if models.none?
-
-    <<~CAPABILITY
-      ### #{capability} (#{models.count})
-
-      #{models_table(models)}
-    CAPABILITY
-  end.join("\n\n")
-end
-
-def generate_modality_sections # rubocop:disable Metrics/PerceivedComplexity
-  sections = []
-
-  vision_models = RubyLLM.models.select { |m| (m.modalities.input || []).include?('image') }
-  if vision_models.any?
-    sections << <<~SECTION
-      ### Vision Models (#{vision_models.count})
-
-      Models that can process images:
-
-      #{models_table(vision_models)}
-    SECTION
-  end
-
-  audio_models = RubyLLM.models.select { |m| (m.modalities.input || []).include?('audio') }
-  if audio_models.any?
-    sections << <<~SECTION
-      ### Audio Input Models (#{audio_models.count})
-
-      Models that can process audio:
-
-      #{models_table(audio_models)}
-    SECTION
-  end
-
-  pdf_models = RubyLLM.models.select { |m| (m.modalities.input || []).include?('pdf') }
-  if pdf_models.any?
-    sections << <<~SECTION
-      ### PDF Models (#{pdf_models.count})
-
-      Models that can process PDF documents:
-
-      #{models_table(pdf_models)}
-    SECTION
-  end
-
-  embedding_models = RubyLLM.models.select { |m| (m.modalities.output || []).include?('embeddings') }
-  if embedding_models.any?
-    sections << <<~SECTION
-      ### Embedding Models (#{embedding_models.count})
-
-      Models that generate embeddings:
-
-      #{models_table(embedding_models)}
-    SECTION
-  end
-
-  sections.join("\n\n")
-end
-
-def models_table(models)
-  return '*No models found*' if models.none?
-
-  headers = ['Model', 'Provider', 'Context', 'Max Output', 'Standard Pricing (per 1M tokens)']
-  alignment = [':--', ':--', '--:', '--:', ':--']
-
-  rows = models.sort_by { |m| [m.provider, m.name] }.map do |model|
-    pricing = standard_pricing_display(model)
-
-    [
-      model.id,
-      model.provider,
-      model.context_window || '-',
-      model.max_output_tokens || '-',
-      pricing
-    ]
-  end
-
-  table = []
-  table << "| #{headers.join(' | ')} |"
-  table << "| #{alignment.join(' | ')} |"
-
-  rows.each do |row|
-    table << "| #{row.join(' | ')} |"
-  end
-
-  table.join("\n")
-end
-
-def standard_pricing_display(model)
-  pricing_data = model.pricing.to_h[:text_tokens]&.dig(:standard) || {}
-
-  if pricing_data.any?
-    parts = []
-
-    parts << "In: $#{format('%.2f', pricing_data[:input_per_million])}" if pricing_data[:input_per_million]
-
-    parts << "Out: $#{format('%.2f', pricing_data[:output_per_million])}" if pricing_data[:output_per_million]
-
-    if pricing_data[:cached_input_per_million]
-      parts << "Cache: $#{format('%.2f', pricing_data[:cached_input_per_million])}"
-    end
-
-    return parts.join(', ') if parts.any?
-  end
-
-  '-'
-end