dify_llm 1.8.2 → 1.9.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 7643dffaaba8bee51ea9d5929f3665be99f4e56fb84f5474fca20f8b8fd71fd5
-  data.tar.gz: 1a5db0333d2df1e09d1b1841e209677d57401137908e58426626844a5e9a11d9
+  metadata.gz: 75dbb53612d3fa2c1089038bcf48fbbc0fe9425d37ffd8fccdfa56337daf97af
+  data.tar.gz: 316d3ef004a7387a6e723a02f8ab09729b167097127f9cb85ce5a864e6e4ef1e
 SHA512:
-  metadata.gz: 305089c41bb76bf36aca07dfd95776b182debe2e059ce75dd34a77627516b2b95603de73780363cf5a9a2096df0109e433fdb420416067761daed700fad7c0b3
-  data.tar.gz: e7478c5771d9d00046d77a2965b801a705e63734360c1871e8865ac9ed634882e529bcf5b241cfeed103e38c87dd009730e6cab92075074b915525525525802e
+  metadata.gz: bca45bf0d49f6e98e9ea00cf2e760537fbff5f861394080f8dec8baa634784289f23779476eedbda5c9d11e4323e2a67f6eadc84144c18cd4adc7d09e3b9cfe7
+  data.tar.gz: a12a868701ae4e70f6f397de290d3bc018da91993950366d956fd8d60f0c37cc8ef89f882a680db396336258107e032f881817224668b81d9d65b3f53c1f44d9

data/README.md

@@ -18,7 +18,7 @@ Battle tested at [<picture><source media="(prefers-color-scheme: dark)" srcset="
 </div>
 
 > [!NOTE]
-> Using RubyLLM in production? [Share your story](https://tally.so/r/3Na02p)! Takes 5 minutes.
+> Using RubyLLM? [Share your story](https://tally.so/r/3Na02p)! Takes 5 minutes.
 
 ---
 
@@ -69,6 +69,11 @@ RubyLLM.paint "a sunset over mountains in watercolor style"
 RubyLLM.embed "Ruby is elegant and expressive"
 ```
 
+```ruby
+# Transcribe audio to text
+RubyLLM.transcribe "meeting.wav"
+```
+
 ```ruby
 # Moderate content for safety
 RubyLLM.moderate "Check if this text is safe"
@@ -107,10 +112,10 @@ response = chat.with_schema(ProductSchema).ask "Analyze this product", with: "pr
 
 * **Chat:** Conversational AI with `RubyLLM.chat`
 * **Vision:** Analyze images and videos
-* **Audio:** Transcribe and understand speech
+* **Audio:** Transcribe and understand speech with `RubyLLM.transcribe`
 * **Documents:** Extract from PDFs, CSVs, JSON, any file type
 * **Image generation:** Create images with `RubyLLM.paint`
-* **Embeddings:** Vector search with `RubyLLM.embed`
+* **Embeddings:** Generate embeddings with `RubyLLM.embed`
 * **Moderation:** Content safety with `RubyLLM.moderate`
 * **Tools:** Let AI call your Ruby methods
 * **Structured output:** JSON schemas that just work

data/lib/generators/ruby_llm/install/templates/create_messages_migration.rb.tt

@@ -3,8 +3,11 @@ class Create<%= message_model_name.gsub('::', '').pluralize %> < ActiveRecord::M
     create_table :<%= message_table_name %> do |t|
       t.string :role, null: false
       t.text :content
+      t.json :content_raw
       t.integer :input_tokens
       t.integer :output_tokens
+      t.integer :cached_tokens
+      t.integer :cache_creation_tokens
       t.timestamps
     end
 

data/lib/generators/ruby_llm/upgrade_to_v1_7/upgrade_to_v1_7_generator.rb

@@ -9,7 +9,7 @@ module RubyLLM
     # Generator to upgrade existing RubyLLM apps to v1.7 with new Rails-like API
     class UpgradeToV17Generator < Rails::Generators::Base
       include Rails::Generators::Migration
-      include RubyLLM::GeneratorHelpers
+      include RubyLLM::Generators::GeneratorHelpers
 
       namespace 'ruby_llm:upgrade_to_v1_7'
       source_root File.expand_path('templates', __dir__)

data/lib/generators/ruby_llm/upgrade_to_v1_9/templates/add_v1_9_message_columns.rb.tt

@@ -0,0 +1,15 @@
+class AddRubyLlmV19Columns < ActiveRecord::Migration<%= migration_version %>
+  def change
+    unless column_exists?(:<%= message_table_name %>, :cached_tokens)
+      add_column :<%= message_table_name %>, :cached_tokens, :integer
+    end
+
+    unless column_exists?(:<%= message_table_name %>, :cache_creation_tokens)
+      add_column :<%= message_table_name %>, :cache_creation_tokens, :integer
+    end
+
+    unless column_exists?(:<%= message_table_name %>, :content_raw)
+      add_column :<%= message_table_name %>, :content_raw, :json
+    end
+  end
+end

data/lib/generators/ruby_llm/upgrade_to_v1_9/upgrade_to_v1_9_generator.rb

@@ -0,0 +1,49 @@
+# frozen_string_literal: true
+
+require 'rails/generators'
+require 'rails/generators/active_record'
+require_relative '../generator_helpers'
+
+module RubyLLM
+  module Generators
+    # Generator to add v1.9 columns (cached tokens + raw content support) to existing apps.
+    class UpgradeToV19Generator < Rails::Generators::Base
+      include Rails::Generators::Migration
+      include RubyLLM::Generators::GeneratorHelpers
+
+      namespace 'ruby_llm:upgrade_to_v1_9'
+      source_root File.expand_path('templates', __dir__)
+
+      argument :model_mappings, type: :array, default: [], banner: 'message:MessageName'
+
+      desc 'Adds cached token columns and raw content storage fields introduced in v1.9.0'
+
+      def self.next_migration_number(dirname)
+        ::ActiveRecord::Generators::Base.next_migration_number(dirname)
+      end
+
+      def create_migration_file
+        parse_model_mappings
+
+        migration_template 'add_v1_9_message_columns.rb.tt',
+                           'db/migrate/add_ruby_llm_v1_9_columns.rb',
+                           migration_version: migration_version,
+                           message_table_name: message_table_name
+      end
+
+      def show_next_steps
+        say_status :success, 'Upgrade prepared!', :green
+        say <<~INSTRUCTIONS
+
+          Next steps:
+          1. Review the generated migration
+          2. Run: rails db:migrate
+          3. Restart your application server
+
+          📚 See the v1.9.0 release notes for details on cached token tracking and raw content support.
+
+        INSTRUCTIONS
+      end
+    end
+  end
+end

data/lib/ruby_llm/active_record/acts_as.rb

@@ -11,22 +11,22 @@ module RubyLLM
         super
         # Monkey-patch Models to use database when ActsAs is active
         RubyLLM::Models.class_eval do
-          def load_models
+          def self.load_models
             read_from_database
           rescue StandardError => e
             RubyLLM.logger.debug "Failed to load models from database: #{e.message}, falling back to JSON"
             read_from_json
           end
 
-          def load_from_database!
-            @models = read_from_database
-          end
-
-          def read_from_database
+          def self.read_from_database
             model_class = RubyLLM.config.model_registry_class
             model_class = model_class.constantize if model_class.is_a?(String)
             model_class.all.map(&:to_llm)
           end
+
+          def load_from_database!
+            @models = self.class.read_from_database
+          end
         end
       end
 

data/lib/ruby_llm/active_record/chat_methods.rb

@@ -174,8 +174,16 @@ module RubyLLM
       end
 
       def create_user_message(content, with: nil)
-        message_record = messages_association.create!(role: :user, content: content)
+        content_text, attachments, content_raw = prepare_content_for_storage(content)
+
+        message_record = messages_association.build(role: :user)
+        message_record.content = content_text
+        message_record.content_raw = content_raw if message_record.respond_to?(:content_raw=)
+        message_record.save!
+
         persist_content(message_record, with) if with.present?
+        persist_content(message_record, attachments) if attachments.present?
+
         message_record
       end
 
@@ -235,28 +243,25 @@ module RubyLLM
         @message = messages_association.create!(role: :assistant, content: '')
       end
 
-      def persist_message_completion(message) # rubocop:disable Metrics/PerceivedComplexity
+      # rubocop:disable Metrics/PerceivedComplexity
+      def persist_message_completion(message)
         return unless message
 
         tool_call_id = find_tool_call_id(message.tool_call_id) if message.tool_call_id
 
         transaction do
-          content = message.content
-          attachments_to_persist = nil
-
-          if content.is_a?(RubyLLM::Content)
-            attachments_to_persist = content.attachments if content.attachments.any?
-            content = content.text
-          elsif content.is_a?(Hash) || content.is_a?(Array)
-            content = content.to_json
-          end
+          content_text, attachments_to_persist, content_raw = prepare_content_for_storage(message.content)
 
           attrs = {
             role: message.role,
-            content: content,
+            content: content_text,
             input_tokens: message.input_tokens,
             output_tokens: message.output_tokens
           }
+          attrs[:cached_tokens] = message.cached_tokens if @message.has_attribute?(:cached_tokens)
+          if @message.has_attribute?(:cache_creation_tokens)
+            attrs[:cache_creation_tokens] = message.cache_creation_tokens
+          end
 
           # Add model association dynamically
           attrs[self.class.model_association_name] = model_association
@@ -266,12 +271,15 @@ module RubyLLM
             attrs[parent_tool_call_assoc.foreign_key] = tool_call_id
           end
 
-          @message.update!(attrs)
+          @message.assign_attributes(attrs)
+          @message.content_raw = content_raw if @message.respond_to?(:content_raw=)
+          @message.save!
 
           persist_content(@message, attachments_to_persist) if attachments_to_persist
           persist_tool_calls(message.tool_calls) if message.tool_calls.present?
         end
       end
+      # rubocop:enable Metrics/PerceivedComplexity
 
       def persist_tool_calls(tool_calls)
         tool_calls.each_value do |tool_call|
@@ -331,6 +339,26 @@ module RubyLLM
         RubyLLM.logger.warn "Failed to process attachment #{source}: #{e.message}"
         nil
       end
+
+      def prepare_content_for_storage(content)
+        attachments = nil
+        content_raw = nil
+        content_text = content
+
+        case content
+        when RubyLLM::Content::Raw
+          content_raw = content.value
+          content_text = nil
+        when RubyLLM::Content
+          attachments = content.attachments if content.attachments.any?
+          content_text = content.text
+        when Hash, Array
+          content_raw = content
+          content_text = nil
+        end
+
+        [content_text, attachments, content_raw]
+      end
     end
   end
 end

data/lib/ruby_llm/active_record/message_methods.rb

@@ -11,6 +11,9 @@ module RubyLLM
       end
 
       def to_llm
+        cached = has_attribute?(:cached_tokens) ? self[:cached_tokens] : nil
+        cache_creation = has_attribute?(:cache_creation_tokens) ? self[:cache_creation_tokens] : nil
+
         RubyLLM::Message.new(
           role: role.to_sym,
           content: extract_content,
@@ -18,6 +21,8 @@ module RubyLLM
           tool_call_id: extract_tool_call_id,
           input_tokens: input_tokens,
           output_tokens: output_tokens,
+          cached_tokens: cached,
+          cache_creation_tokens: cache_creation,
           model_id: model_association&.model_id
         )
       end
@@ -42,9 +47,13 @@ module RubyLLM
       end
 
       def extract_content
-        return content unless respond_to?(:attachments) && attachments.attached?
+        return RubyLLM::Content::Raw.new(content_raw) if has_attribute?(:content_raw) && content_raw.present?
+
+        content_value = self[:content]
+
+        return content_value unless respond_to?(:attachments) && attachments.attached?
 
-        RubyLLM::Content.new(content).tap do |content_obj|
+        RubyLLM::Content.new(content_value).tap do |content_obj|
           @_tempfiles = []
 
           attachments.each do |attachment|

data/lib/ruby_llm/active_record/model_methods.rb

@@ -77,7 +77,7 @@ module RubyLLM
       delegate :supports?, :supports_vision?, :supports_functions?, :type,
                :input_price_per_million, :output_price_per_million,
                :function_calling?, :structured_output?, :batch?,
-               :reasoning?, :citations?, :streaming?,
+               :reasoning?, :citations?, :streaming?, :provider_class,
                to: :to_llm
     end
   end

data/lib/ruby_llm/aliases.json

@@ -8,6 +8,9 @@
     "openrouter": "anthropic/claude-3.5-haiku",
     "bedrock": "anthropic.claude-3-5-haiku-20241022-v1:0"
   },
+  "claude-3-5-haiku-latest": {
+    "anthropic": "claude-3-5-haiku-latest"
+  },
   "claude-3-5-sonnet": {
     "anthropic": "claude-3-5-sonnet-20241022",
     "openrouter": "anthropic/claude-3.5-sonnet",
@@ -18,6 +21,9 @@
     "openrouter": "anthropic/claude-3.7-sonnet",
     "bedrock": "us.anthropic.claude-3-7-sonnet-20250219-v1:0"
   },
+  "claude-3-7-sonnet-latest": {
+    "anthropic": "claude-3-7-sonnet-latest"
+  },
   "claude-3-haiku": {
     "anthropic": "claude-3-haiku-20240307",
     "openrouter": "anthropic/claude-3-haiku",
@@ -31,11 +37,19 @@
   "claude-3-sonnet": {
     "bedrock": "anthropic.claude-3-sonnet-20240229-v1:0"
   },
+  "claude-haiku-4-5": {
+    "anthropic": "claude-haiku-4-5-20251001",
+    "openrouter": "anthropic/claude-haiku-4.5",
+    "bedrock": "us.anthropic.claude-haiku-4-5-20251001-v1:0"
+  },
   "claude-opus-4": {
     "anthropic": "claude-opus-4-20250514",
     "openrouter": "anthropic/claude-opus-4",
     "bedrock": "us.anthropic.claude-opus-4-1-20250805-v1:0"
   },
+  "claude-opus-4-0": {
+    "anthropic": "claude-opus-4-0"
+  },
   "claude-opus-4-1": {
     "anthropic": "claude-opus-4-1-20250805",
     "openrouter": "anthropic/claude-opus-4.1",
@@ -46,30 +60,18 @@
     "openrouter": "anthropic/claude-sonnet-4",
     "bedrock": "us.anthropic.claude-sonnet-4-20250514-v1:0"
   },
+  "claude-sonnet-4-0": {
+    "anthropic": "claude-sonnet-4-0"
+  },
+  "claude-sonnet-4-5": {
+    "anthropic": "claude-sonnet-4-5-20250929",
+    "openrouter": "anthropic/claude-sonnet-4.5",
+    "bedrock": "us.anthropic.claude-sonnet-4-5-20250929-v1:0"
+  },
   "deepseek-chat": {
     "deepseek": "deepseek-chat",
     "openrouter": "deepseek/deepseek-chat"
   },
-  "gemini-1.5-flash": {
-    "gemini": "gemini-1.5-flash",
-    "vertexai": "gemini-1.5-flash"
-  },
-  "gemini-1.5-flash-002": {
-    "gemini": "gemini-1.5-flash-002",
-    "vertexai": "gemini-1.5-flash-002"
-  },
-  "gemini-1.5-flash-8b": {
-    "gemini": "gemini-1.5-flash-8b",
-    "vertexai": "gemini-1.5-flash-8b"
-  },
-  "gemini-1.5-pro": {
-    "gemini": "gemini-1.5-pro",
-    "vertexai": "gemini-1.5-pro"
-  },
-  "gemini-1.5-pro-002": {
-    "gemini": "gemini-1.5-pro-002",
-    "vertexai": "gemini-1.5-pro-002"
-  },
   "gemini-2.0-flash": {
     "gemini": "gemini-2.0-flash",
     "vertexai": "gemini-2.0-flash"
@@ -93,6 +95,10 @@
     "openrouter": "google/gemini-2.5-flash",
     "vertexai": "gemini-2.5-flash"
   },
+  "gemini-2.5-flash-image": {
+    "gemini": "gemini-2.5-flash-image",
+    "openrouter": "google/gemini-2.5-flash-image"
+  },
   "gemini-2.5-flash-image-preview": {
     "gemini": "gemini-2.5-flash-image-preview",
     "openrouter": "google/gemini-2.5-flash-image-preview"
@@ -106,6 +112,14 @@
     "gemini": "gemini-2.5-flash-lite-preview-06-17",
     "openrouter": "google/gemini-2.5-flash-lite-preview-06-17"
   },
+  "gemini-2.5-flash-lite-preview-09-2025": {
+    "gemini": "gemini-2.5-flash-lite-preview-09-2025",
+    "openrouter": "google/gemini-2.5-flash-lite-preview-09-2025"
+  },
+  "gemini-2.5-flash-preview-09-2025": {
+    "gemini": "gemini-2.5-flash-preview-09-2025",
+    "openrouter": "google/gemini-2.5-flash-preview-09-2025"
+  },
   "gemini-2.5-pro": {
     "gemini": "gemini-2.5-pro",
     "openrouter": "google/gemini-2.5-pro",
@@ -219,6 +233,10 @@
     "openai": "gpt-5",
     "openrouter": "openai/gpt-5"
   },
+  "gpt-5-codex": {
+    "openai": "gpt-5-codex",
+    "openrouter": "openai/gpt-5-codex"
+  },
   "gpt-5-mini": {
     "openai": "gpt-5-mini",
     "openrouter": "openai/gpt-5-mini"
@@ -227,6 +245,22 @@
     "openai": "gpt-5-nano",
     "openrouter": "openai/gpt-5-nano"
   },
+  "gpt-5-pro": {
+    "openai": "gpt-5-pro",
+    "openrouter": "openai/gpt-5-pro"
+  },
+  "gpt-oss-120b": {
+    "openai": "gpt-oss-120b",
+    "openrouter": "openai/gpt-oss-120b"
+  },
+  "gpt-oss-20b": {
+    "openai": "gpt-oss-20b",
+    "openrouter": "openai/gpt-oss-20b"
+  },
+  "imagen-4.0-generate-001": {
+    "gemini": "imagen-4.0-generate-001",
+    "vertexai": "imagen-4.0-generate-001"
+  },
   "o1": {
     "openai": "o1",
     "openrouter": "openai/o1"
@@ -247,6 +281,10 @@
     "openai": "o3",
     "openrouter": "openai/o3"
   },
+  "o3-deep-research": {
+    "openai": "o3-deep-research",
+    "openrouter": "openai/o3-deep-research"
+  },
   "o3-mini": {
     "openai": "o3-mini",
     "openrouter": "openai/o3-mini"
@@ -259,6 +297,10 @@
     "openai": "o4-mini",
     "openrouter": "openai/o4-mini"
   },
+  "o4-mini-deep-research": {
+    "openai": "o4-mini-deep-research",
+    "openrouter": "openai/o4-mini-deep-research"
+  },
   "text-embedding-004": {
     "gemini": "text-embedding-004",
     "vertexai": "text-embedding-004"

data/lib/ruby_llm/attachment.rb

@@ -71,6 +71,14 @@ module RubyLLM
       Base64.strict_encode64(content)
     end
 
+    def save(path)
+      return unless io_like?
+
+      File.open(path, 'w') do |f|
+        f.puts(@source.read)
+      end
+    end
+
     def for_llm
       case type
       when :text

data/lib/ruby_llm/chat.rb

@@ -31,7 +31,7 @@ module RubyLLM
     end
 
     def ask(message = nil, with: nil, &)
-      add_message role: :user, content: Content.new(message, with)
+      add_message role: :user, content: build_content(message, with)
       complete(&)
     end
 
@@ -193,7 +193,8 @@ module RubyLLM
         @on[:tool_call]&.call(tool_call)
         result = execute_tool tool_call
         @on[:tool_result]&.call(result)
-        content = result.is_a?(Content) ? result : result.to_s
+        tool_payload = result.is_a?(Tool::Halt) ? result.content : result
+        content = content_like?(tool_payload) ? tool_payload : tool_payload.to_s
         message = add_message role: :tool, content:, tool_call_id: tool_call.id
         @on[:end_message]&.call(message)
 
@@ -208,5 +209,15 @@ module RubyLLM
       args = tool_call.arguments
       tool.call(args)
     end
+
+    def build_content(message, attachments)
+      return message if content_like?(message)
+
+      Content.new(message, attachments)
+    end
+
+    def content_like?(object)
+      object.is_a?(Content) || object.is_a?(Content::Raw)
+    end
   end
 end

data/lib/ruby_llm/configuration.rb

@@ -10,6 +10,7 @@ module RubyLLM
                   :openai_use_system_role,
                   :anthropic_api_key,
                   :gemini_api_key,
+                  :gemini_api_base,
                   :vertexai_project_id,
                   :vertexai_location,
                   :deepseek_api_key,
@@ -31,7 +32,9 @@ module RubyLLM
                   :default_embedding_model,
                   :default_moderation_model,
                   :default_image_model,
+                  :default_transcription_model,
                   # Model registry
+                  :model_registry_file,
                   :model_registry_class,
                   # Rails integration
                   :use_new_acts_as,
@@ -49,7 +52,7 @@ module RubyLLM
                   :log_stream_debug
 
     def initialize
-      @request_timeout = 120
+      @request_timeout = 300
       @max_retries = 3
       @retry_interval = 0.1
       @retry_backoff_factor = 2
@@ -60,7 +63,9 @@ module RubyLLM
       @default_embedding_model = 'text-embedding-3-small'
       @default_moderation_model = 'omni-moderation-latest'
       @default_image_model = 'gpt-image-1'
+      @default_transcription_model = 'whisper-1'
 
+      @model_registry_file = File.expand_path('models.json', __dir__)
       @model_registry_class = 'Model'
       @use_new_acts_as = false
 

data/lib/ruby_llm/connection.rb

@@ -34,8 +34,7 @@ module RubyLLM
     end
 
     def post(url, payload, &)
-      body = payload.is_a?(Hash) ? JSON.generate(payload, ascii_only: false) : payload
-      @connection.post url, body do |req|
+      @connection.post url, payload do |req|
         req.headers.merge! @provider.headers if @provider.respond_to?(:headers)
         yield req if block_given?
       end
@@ -77,7 +76,7 @@ module RubyLLM
                        errors: true,
                        headers: false,
                        log_level: :debug do |logger|
-        logger.filter(%r{[A-Za-z0-9+/=]{100,}}, 'data":"[BASE64 DATA]"')
+        logger.filter(%r{[A-Za-z0-9+/=]{100,}}, '[BASE64 DATA]')
         logger.filter(/[-\d.e,\s]{100,}/, '[EMBEDDINGS ARRAY]')
       end
     end
@@ -94,6 +93,7 @@ module RubyLLM
     end
 
     def setup_middleware(faraday)
+      faraday.request :multipart
       faraday.request :json
       faraday.response :json
       faraday.adapter :net_http

data/lib/ruby_llm/content.rb

@@ -48,3 +48,26 @@ module RubyLLM
     end
   end
 end
+
+module RubyLLM
+  class Content
+    # Represents provider-specific payloads that should bypass RubyLLM formatting.
+    class Raw
+      attr_reader :value
+
+      def initialize(value)
+        raise ArgumentError, 'Raw content payload cannot be nil' if value.nil?
+
+        @value = value
+      end
+
+      def format
+        @value
+      end
+
+      def to_h
+        @value
+      end
+    end
+  end
+end

data/lib/ruby_llm/message.rb

@@ -5,18 +5,21 @@ module RubyLLM
   class Message
     ROLES = %i[system user assistant tool].freeze
 
-    attr_reader :role, :tool_calls, :tool_call_id, :input_tokens, :output_tokens, :model_id, :raw, :conversation_id
+    attr_reader :role, :model_id, :tool_calls, :tool_call_id, :input_tokens, :output_tokens,
+                :cached_tokens, :cache_creation_tokens, :raw, :conversation_id
     attr_writer :content
 
     def initialize(options = {})
       @role = options.fetch(:role).to_sym
       @content = normalize_content(options.fetch(:content))
+      @model_id = options[:model_id]
       @tool_calls = options[:tool_calls]
+      @tool_call_id = options[:tool_call_id]
+      @conversation_id = options[:conversation_id]
       @input_tokens = options[:input_tokens]
       @output_tokens = options[:output_tokens]
-      @model_id = options[:model_id]
-      @conversation_id = options[:conversation_id]
-      @tool_call_id = options[:tool_call_id]
+      @cached_tokens = options[:cached_tokens]
+      @cache_creation_tokens = options[:cache_creation_tokens]
       @raw = options[:raw]
 
       ensure_valid_role
@@ -46,12 +49,14 @@ module RubyLLM
       {
         role: role,
         content: content,
+        model_id: model_id,
         tool_calls: tool_calls,
         tool_call_id: tool_call_id,
+        conversation_id: conversation_id,
         input_tokens: input_tokens,
         output_tokens: output_tokens,
-        conversation_id: conversation_id,
-        model_id: model_id
+        cached_tokens: cached_tokens,
+        cache_creation_tokens: cache_creation_tokens
       }.compact
     end
 

data/lib/ruby_llm/model/info.rb

@@ -72,6 +72,10 @@ module RubyLLM
         pricing.text_tokens.output
       end
 
+      def provider_class
+        RubyLLM::Provider.resolve provider
+      end
+
       def type # rubocop:disable Metrics/PerceivedComplexity
         if modalities.output.include?('embeddings') && !modalities.output.include?('text')
           'embedding'