Shorter model name params (#840)

* rename chat_completion_model_name -> chat_model and completion_model_name -> completion_model * update readme * update changelog * bump version * update gemfile.lock * fix nits * rename embeddings_model_name -> embed_model * rename completion_model -> complete_model * update changelog * changelog md formatting * Update Gemfile.lock * Update CHANGELOG.md * Renaming the parameters * Update version.rb --------- Co-authored-by: Andrei Bondarev <[email protected]> Co-authored-by: Andrei Bondarev <[email protected]>
patterns-ai-core · Oct 23, 2024 · cf2bafd · cf2bafd
1 parent 8091e54
commit cf2bafd
Show file tree

Hide file tree

Showing 26 changed files with 106 additions and 103 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -10,6 +10,9 @@
 - [SECURITY]: A change which fixes a security vulnerability.
 
 ## [Unreleased]
+- [BREAKING] [https://github.com/patterns-ai-core/langchainrb/pull/840] Rename `chat_completion_model_name` parameter to `chat_model` in Langchain::LLM parameters.
+- [BREAKING] [https://github.com/patterns-ai-core/langchainrb/pull/840] Rename `completion_model_name` parameter to `completion_model` in Langchain::LLM parameters.
+- [BREAKING] [https://github.com/patterns-ai-core/langchainrb/pull/840] Rename `embeddings_model_name` parameter to `embedding_model` in Langchain::LLM parameters.
 - [BUGFIX] [https://github.com/patterns-ai-core/langchainrb/pull/850/] Fix MistralAIMessage to handle "Tool" Output
 - [BUGFIX] [https://github.com/patterns-ai-core/langchainrb/pull/837] Fix bug when tool functions with no input variables are used with Langchain::LLM::Anthropic
 - [BUGFIX] [https://github.com/patterns-ai-core/langchainrb/pull/836] Fix bug when assistant.instructions = nil did not remove the system message

diff --git a/README.md b/README.md
@@ -86,7 +86,7 @@ Most LLM classes can be initialized with an API key and optional default options
 ```ruby
 llm = Langchain::LLM::OpenAI.new(
   api_key: ENV["OPENAI_API_KEY"],
-  default_options: { temperature: 0.7, chat_completion_model_name: "gpt-4o" }
+  default_options: { temperature: 0.7, chat_model: "gpt-4o" }
 )
 ```
 

diff --git a/examples/openai_qdrant_function_calls.rb b/examples/openai_qdrant_function_calls.rb
@@ -21,7 +21,7 @@
 openai = Langchain::LLM::OpenAI.new(
   api_key: ENV["OPENAI_API_KEY"],
   default_options: {
-    chat_completion_model_name: "gpt-3.5-turbo-16k"
+    chat_model: "gpt-3.5-turbo-16k"
   }
 )
 

diff --git a/lib/langchain/llm/anthropic.rb b/lib/langchain/llm/anthropic.rb
@@ -13,24 +13,24 @@ module Langchain::LLM
   class Anthropic < Base
     DEFAULTS = {
       temperature: 0.0,
-      completion_model_name: "claude-2.1",
-      chat_completion_model_name: "claude-3-5-sonnet-20240620",
+      completion_model: "claude-2.1",
+      chat_model: "claude-3-5-sonnet-20240620",
       max_tokens_to_sample: 256
     }.freeze
 
     # Initialize an Anthropic LLM instance
     #
     # @param api_key [String] The API key to use
     # @param llm_options [Hash] Options to pass to the Anthropic client
-    # @param default_options [Hash] Default options to use on every call to LLM, e.g.: { temperature:, completion_model_name:, chat_completion_model_name:, max_tokens_to_sample: }
+    # @param default_options [Hash] Default options to use on every call to LLM, e.g.: { temperature:, completion_model:, chat_model:, max_tokens_to_sample: }
     # @return [Langchain::LLM::Anthropic] Langchain::LLM::Anthropic instance
     def initialize(api_key:, llm_options: {}, default_options: {})
       depends_on "anthropic"
 
       @client = ::Anthropic::Client.new(access_token: api_key, **llm_options)
       @defaults = DEFAULTS.merge(default_options)
       chat_parameters.update(
-        model: {default: @defaults[:chat_completion_model_name]},
+        model: {default: @defaults[:chat_model]},
         temperature: {default: @defaults[:temperature]},
         max_tokens: {default: @defaults[:max_tokens_to_sample]},
         metadata: {},
@@ -54,7 +54,7 @@ def initialize(api_key:, llm_options: {}, default_options: {})
     # @return [Langchain::LLM::AnthropicResponse] The completion
     def complete(
       prompt:,
-      model: @defaults[:completion_model_name],
+      model: @defaults[:completion_model],
       max_tokens_to_sample: @defaults[:max_tokens_to_sample],
       stop_sequences: nil,
       temperature: @defaults[:temperature],

diff --git a/lib/langchain/llm/aws_bedrock.rb b/lib/langchain/llm/aws_bedrock.rb
@@ -11,9 +11,9 @@ module Langchain::LLM
   #
   class AwsBedrock < Base
     DEFAULTS = {
-      chat_completion_model_name: "anthropic.claude-v2",
-      completion_model_name: "anthropic.claude-v2",
-      embeddings_model_name: "amazon.titan-embed-text-v1",
+      chat_model: "anthropic.claude-v2",
+      completion_model: "anthropic.claude-v2",
+      embedding_model: "amazon.titan-embed-text-v1",
       max_tokens_to_sample: 300,
       temperature: 1,
       top_k: 250,
@@ -60,7 +60,7 @@ def initialize(aws_client_options: {}, default_options: {})
       @defaults = DEFAULTS.merge(default_options)
 
       chat_parameters.update(
-        model: {default: @defaults[:chat_completion_model_name]},
+        model: {default: @defaults[:chat_model]},
         temperature: {},
         max_tokens: {default: @defaults[:max_tokens_to_sample]},
         metadata: {},
@@ -84,7 +84,7 @@ def embed(text:, **params)
       parameters = compose_embedding_parameters params.merge(text:)
 
       response = client.invoke_model({
-        model_id: @defaults[:embeddings_model_name],
+        model_id: @defaults[:embedding_model],
         body: parameters.to_json,
         content_type: "application/json",
         accept: "application/json"
@@ -103,14 +103,14 @@ def embed(text:, **params)
     def complete(prompt:, **params)
       raise "Completion provider #{completion_provider} is not supported." unless SUPPORTED_COMPLETION_PROVIDERS.include?(completion_provider)
 
-      raise "Model #{@defaults[:completion_model_name]} only supports #chat." if @defaults[:completion_model_name].include?("claude-3")
+      raise "Model #{@defaults[:completion_model]} only supports #chat." if @defaults[:completion_model].include?("claude-3")
 
       parameters = compose_parameters params
 
       parameters[:prompt] = wrap_prompt prompt
 
       response = client.invoke_model({
-        model_id: @defaults[:completion_model_name],
+        model_id: @defaults[:completion_model],
         body: parameters.to_json,
         content_type: "application/json",
         accept: "application/json"
@@ -126,7 +126,7 @@ def complete(prompt:, **params)
     # @param [Hash] params unified chat parmeters from [Langchain::LLM::Parameters::Chat::SCHEMA]
     # @option params [Array<String>] :messages The messages to generate a completion for
     # @option params [String] :system The system prompt to provide instructions
-    # @option params [String] :model The model to use for completion defaults to @defaults[:chat_completion_model_name]
+    # @option params [String] :model The model to use for completion defaults to @defaults[:chat_model]
     # @option params [Integer] :max_tokens The maximum number of tokens to generate defaults to @defaults[:max_tokens_to_sample]
     # @option params [Array<String>] :stop The stop sequences to use for completion
     # @option params [Array<String>] :stop_sequences The stop sequences to use for completion
@@ -175,11 +175,11 @@ def chat(params = {}, &block)
     private
 
     def completion_provider
-      @defaults[:completion_model_name].split(".").first.to_sym
+      @defaults[:completion_model].split(".").first.to_sym
     end
 
     def embedding_provider
-      @defaults[:embeddings_model_name].split(".").first.to_sym
+      @defaults[:embedding_model].split(".").first.to_sym
     end
 
     def wrap_prompt(prompt)

diff --git a/lib/langchain/llm/azure.rb b/lib/langchain/llm/azure.rb
@@ -33,7 +33,7 @@ def initialize(
       )
       @defaults = DEFAULTS.merge(default_options)
       chat_parameters.update(
-        model: {default: @defaults[:chat_completion_model_name]},
+        model: {default: @defaults[:chat_model]},
         logprobs: {},
         top_logprobs: {},
         n: {default: @defaults[:n]},

diff --git a/lib/langchain/llm/base.rb b/lib/langchain/llm/base.rb
@@ -34,7 +34,7 @@ def default_dimension
       default_dimensions
     end
 
-    # Returns the number of vector dimensions used by DEFAULTS[:chat_completion_model_name]
+    # Returns the number of vector dimensions used by DEFAULTS[:chat_model]
     #
     # @return [Integer] Vector dimensions
     def default_dimensions

diff --git a/lib/langchain/llm/cohere.rb b/lib/langchain/llm/cohere.rb
@@ -13,9 +13,9 @@ module Langchain::LLM
   class Cohere < Base
     DEFAULTS = {
       temperature: 0.0,
-      completion_model_name: "command",
-      chat_completion_model_name: "command-r-plus",
-      embeddings_model_name: "small",
+      completion_model: "command",
+      chat_model: "command-r-plus",
+      embedding_model: "small",
       dimensions: 1024,
       truncate: "START"
     }.freeze
@@ -26,7 +26,7 @@ def initialize(api_key:, default_options: {})
       @client = ::Cohere::Client.new(api_key: api_key)
       @defaults = DEFAULTS.merge(default_options)
       chat_parameters.update(
-        model: {default: @defaults[:chat_completion_model_name]},
+        model: {default: @defaults[:chat_model]},
         temperature: {default: @defaults[:temperature]},
         response_format: {default: @defaults[:response_format]}
       )
@@ -48,10 +48,10 @@ def initialize(api_key:, default_options: {})
     def embed(text:)
       response = client.embed(
         texts: [text],
-        model: @defaults[:embeddings_model_name]
+        model: @defaults[:embedding_model]
       )
 
-      Langchain::LLM::CohereResponse.new response, model: @defaults[:embeddings_model_name]
+      Langchain::LLM::CohereResponse.new response, model: @defaults[:embedding_model]
     end
 
     #
@@ -65,7 +65,7 @@ def complete(prompt:, **params)
       default_params = {
         prompt: prompt,
         temperature: @defaults[:temperature],
-        model: @defaults[:completion_model_name],
+        model: @defaults[:completion_model],
         truncate: @defaults[:truncate]
       }
 
@@ -76,7 +76,7 @@ def complete(prompt:, **params)
       default_params.merge!(params)
 
       response = client.generate(**default_params)
-      Langchain::LLM::CohereResponse.new response, model: @defaults[:completion_model_name]
+      Langchain::LLM::CohereResponse.new response, model: @defaults[:completion_model]
     end
 
     # Generate a chat completion for given messages

diff --git a/lib/langchain/llm/google_gemini.rb b/lib/langchain/llm/google_gemini.rb
@@ -5,8 +5,8 @@ module Langchain::LLM
   #     llm = Langchain::LLM::GoogleGemini.new(api_key: ENV['GOOGLE_GEMINI_API_KEY'])
   class GoogleGemini < Base
     DEFAULTS = {
-      chat_completion_model_name: "gemini-1.5-pro-latest",
-      embeddings_model_name: "text-embedding-004",
+      chat_model: "gemini-1.5-pro-latest",
+      embedding_model: "text-embedding-004",
       temperature: 0.0
     }
 
@@ -17,7 +17,7 @@ def initialize(api_key:, default_options: {})
       @defaults = DEFAULTS.merge(default_options)
 
       chat_parameters.update(
-        model: {default: @defaults[:chat_completion_model_name]},
+        model: {default: @defaults[:chat_model]},
         temperature: {default: @defaults[:temperature]},
         generation_config: {default: nil},
         safety_settings: {default: @defaults[:safety_settings]}
@@ -72,9 +72,8 @@ def chat(params = {})
 
     def embed(
       text:,
-      model: @defaults[:embeddings_model_name]
+      model: @defaults[:embedding_model]
     )
-
       params = {
         content: {
           parts: [

diff --git a/lib/langchain/llm/google_vertex_ai.rb b/lib/langchain/llm/google_vertex_ai.rb
@@ -17,8 +17,8 @@ class GoogleVertexAI < Base
       top_p: 0.8,
       top_k: 40,
       dimensions: 768,
-      embeddings_model_name: "textembedding-gecko",
-      chat_completion_model_name: "gemini-1.0-pro"
+      embedding_model: "textembedding-gecko",
+      chat_model: "gemini-1.0-pro"
     }.freeze
 
     # Google Cloud has a project id and a specific region of deployment.
@@ -38,7 +38,7 @@ def initialize(project_id:, region:, default_options: {})
       @defaults = DEFAULTS.merge(default_options)
 
       chat_parameters.update(
-        model: {default: @defaults[:chat_completion_model_name]},
+        model: {default: @defaults[:chat_model]},
         temperature: {default: @defaults[:temperature]},
         safety_settings: {default: @defaults[:safety_settings]}
       )
@@ -58,7 +58,7 @@ def initialize(project_id:, region:, default_options: {})
     #
     def embed(
       text:,
-      model: @defaults[:embeddings_model_name]
+      model: @defaults[:embedding_model]
     )
       params = {instances: [{content: text}]}
 

diff --git a/lib/langchain/llm/hugging_face.rb b/lib/langchain/llm/hugging_face.rb
@@ -12,7 +12,7 @@ module Langchain::LLM
   #
   class HuggingFace < Base
     DEFAULTS = {
-      embeddings_model_name: "sentence-transformers/all-MiniLM-L6-v2"
+      embedding_model: "sentence-transformers/all-MiniLM-L6-v2"
     }.freeze
 
     EMBEDDING_SIZES = {
@@ -36,7 +36,7 @@ def initialize(api_key:, default_options: {})
     def default_dimensions
       # since Huggin Face can run multiple models, look it up or generate an embedding and return the size
       @default_dimensions ||= @defaults[:dimensions] ||
-        EMBEDDING_SIZES.fetch(@defaults[:embeddings_model_name].to_sym) do
+        EMBEDDING_SIZES.fetch(@defaults[:embedding_model].to_sym) do
           embed(text: "test").embedding.size
         end
     end
@@ -50,9 +50,9 @@ def default_dimensions
     def embed(text:)
       response = client.embedding(
         input: text,
-        model: @defaults[:embeddings_model_name]
+        model: @defaults[:embedding_model]
       )
-      Langchain::LLM::HuggingFaceResponse.new(response, model: @defaults[:embeddings_model_name])
+      Langchain::LLM::HuggingFaceResponse.new(response, model: @defaults[:embedding_model])
     end
   end
 end
diff --git a/lib/langchain/llm/mistral_ai.rb b/lib/langchain/llm/mistral_ai.rb
@@ -8,8 +8,8 @@ module Langchain::LLM
   #    llm = Langchain::LLM::MistralAI.new(api_key: ENV["MISTRAL_AI_API_KEY"])
   class MistralAI < Base
     DEFAULTS = {
-      chat_completion_model_name: "mistral-large-latest",
-      embeddings_model_name: "mistral-embed"
+      chat_model: "mistral-large-latest",
+      embedding_model: "mistral-embed"
     }.freeze
 
     attr_reader :defaults
@@ -24,7 +24,7 @@ def initialize(api_key:, default_options: {})
 
       @defaults = DEFAULTS.merge(default_options)
       chat_parameters.update(
-        model: {default: @defaults[:chat_completion_model_name]},
+        model: {default: @defaults[:chat_model]},
         n: {default: @defaults[:n]},
         safe_prompt: {},
         temperature: {default: @defaults[:temperature]},
@@ -44,7 +44,7 @@ def chat(params = {})
 
     def embed(
       text:,
-      model: defaults[:embeddings_model_name],
+      model: defaults[:embedding_model],
       encoding_format: nil
     )
       params = {

diff --git a/lib/langchain/llm/ollama.rb b/lib/langchain/llm/ollama.rb
@@ -12,9 +12,9 @@ class Ollama < Base
 
     DEFAULTS = {
       temperature: 0.0,
-      completion_model_name: "llama3.1",
-      embeddings_model_name: "llama3.1",
-      chat_completion_model_name: "llama3.1"
+      completion_model: "llama3.1",
+      embedding_model: "llama3.1",
+      chat_model: "llama3.1"
     }.freeze
 
     EMBEDDING_SIZES = {
@@ -41,7 +41,7 @@ def initialize(url: "http://localhost:11434", api_key: nil, default_options: {})
       @api_key = api_key
       @defaults = DEFAULTS.merge(default_options)
       chat_parameters.update(
-        model: {default: @defaults[:chat_completion_model_name]},
+        model: {default: @defaults[:chat_model]},
         temperature: {default: @defaults[:temperature]},
         template: {},
         stream: {default: false},
@@ -55,7 +55,7 @@ def initialize(url: "http://localhost:11434", api_key: nil, default_options: {})
     def default_dimensions
       # since Ollama can run multiple models, look it up or generate an embedding and return the size
       @default_dimensions ||=
-        EMBEDDING_SIZES.fetch(defaults[:embeddings_model_name].to_sym) do
+        EMBEDDING_SIZES.fetch(defaults[:embedding_model].to_sym) do
           embed(text: "test").embedding.size
         end
     end
@@ -77,7 +77,7 @@ def default_dimensions
     #
     def complete(
       prompt:,
-      model: defaults[:completion_model_name],
+      model: defaults[:completion_model],
       images: nil,
       format: nil,
       system: nil,
@@ -199,7 +199,7 @@ def chat(messages:, model: nil, **params, &block)
     #
     def embed(
       text:,
-      model: defaults[:embeddings_model_name],
+      model: defaults[:embedding_model],
       mirostat: nil,
       mirostat_eta: nil,
       mirostat_tau: nil,