omniai 2.7.0 → 2.8.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 488be723c0a59ca0095eb3a1f64081cceff14ae982208e598da5058ee0ea0c4b
-  data.tar.gz: 47805f82344429912eaa2e0f00299cc95868d35d04a6d424f698b719bf9d0fe8
+  metadata.gz: a80f4da0021307bce467f92f209309aaea62060e93fbb4201a7fc059b0376cea
+  data.tar.gz: 329e4a51e2e8b1d66ed5eca4bf7979acc1111f720d62fdfb638d850ca9ff53d8
 SHA512:
-  metadata.gz: 1c0d66e74c7a0e1eb0296e371b0339d44cedbb7a747c2beeb7cb0547303140d0a6550fd8d7fc6629c09804c57cfcf780dec38f01f6c8aac49004ab28e9968872
-  data.tar.gz: aeb129f452a6ea0a9c2e33a25afbe5f8be1e7b27445fb58399556e7bdd0475b0840b36465c6b33edbdf857c6d943c10824b0cc21b7e6a917d161829e4ca033a6
+  metadata.gz: 5ce5f9abf61e35856183ebfdf344f41d2961f67140abdd19b48464203a10358e368b606ec8d775e32a515e202c87b69c205a3dc688743ddb072ef70962753948
+  data.tar.gz: f517cf093a2304dd4d6562bdf8a96c78480523063e9ff21463b08b2214064efdd65782af50ef070c836502e54dccce37d66f3636027c41409a62dff1abd2eb83

data/README.md

@@ -10,13 +10,14 @@ OmniAI provides a unified Ruby API for integrating with multiple AI providers, i
 
 - [OmniAI::Anthropic](https://github.com/ksylvest/omniai-anthropic)
 - [OmniAI::DeepSeek](https://github.com/ksylvest/omniai-deepseek)
+- [OmniAI::Llama](https://github.com/ksylvest/omniai-llama)
 - [OmniAI::Google](https://github.com/ksylvest/omniai-google)
 - [OmniAI::Mistral](https://github.com/ksylvest/omniai-mistral)
 - [OmniAI::OpenAI](https://github.com/ksylvest/omniai-openai)
 
-## Examples
+## 📄 Examples
 
-### Example #1: [Chat w/ Text](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_text)
+### Example #1: [💬 Chat w/ Text](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_text)
 
 This example demonstrates using `OmniAI` with **Anthropic** to ask for a joke. The response is parsed and printed.
 
@@ -32,7 +33,7 @@ puts client.chat("Tell me a joke").text
 Why don't scientists trust atoms? Because they make up everything!
 ```
 
-### Example #2: [Chat w/ Prompt](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_prompt)
+### Example #2: [💬 Chat w/ Prompt](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_prompt)
 
 This example demonstrates using `OmniAI` with **Mistral** to ask for the fastest animal. It includes a system and user message in the prompt. The response is streamed in real time.
 
@@ -52,7 +53,7 @@ end
 **French**: Le faucon pèlerin est généralement considéré comme l'animal le plus rapide, atteignant des vitesses de plus de 390 km/h.
 ```
 
-### Example #3: [Chat w/ Vision](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_vision)
+### Example #3: [💬 Chat w/ Vision](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_vision)
 
 This example demonstrates using `OmniAI` with **OpenAI** to prompt a “biologist” for an analysis of photos, identifying the animals within each one. A system and user message are provided, and the response is streamed in real time.
 
@@ -79,7 +80,7 @@ The first photo is of a cat, *Felis Catus*.
 The second photo is of a dog, *Canis Familiaris*.
 ```
 
-### Example #4: [Chat w/ Tools](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_tools)
+### Example #4: [💬 Chat w/ Tools](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_tools)
 
 This example demonstrates using `OmniAI` with **Google** to ask for the weather. A tool “Weather” is provided. The tool accepts a location and unit (Celsius or Fahrenheit) then calculates the weather. The LLM makes multiple tool-call requests and is automatically provided with a tool-call response prior to streaming in real-time the result.
 
@@ -149,7 +150,7 @@ The weather is 24° Celsius in London and 42° Fahrenheit in Madrid.
 
 _For a set of pre-built tools for interacting with browsers, databases, docker, and more try the [OmniAI::Tools](https://omniai-tools.ksylvest.com/) project._
 
-### Example #5: [Chat w/ History](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_history)
+### Example #5: [💬 Chat w/ History](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_history)
 
 Tracking a prompt history over multiple user and assistant messages is especially helpful when building an agent like conversation experience. A prompt can be used to track this back-and-forth conversation:
 
@@ -176,7 +177,7 @@ loop do
 end
 ```
 
-### Example #6 [Chat w/ Schema](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_schema)
+### Example #6 [💬 Chat w/ Schema](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_schema)
 
 Requesting structured data back from an LLM is possible by defining a schema, then passing the schema into the chat. The following example defines a structured schema using `OmniAI::Schema` to model a `Contact`. The results of the LLM call are then parsed using the schema to ensure all types are correct.
 
@@ -228,27 +229,30 @@ puts format.parse(response.text)
 }
 ```
 
-### Example #7: [Chat w/ CLI](https://github.com/ksylvest/omniai/blob/main/examples/chat_with_cli)
+### Example #7: [🐚 CLI](https://github.com/ksylvest/omniai/blob/main/examples/cli)
 
 The `OmniAI` gem also ships with a CLI to simplify quick tests.
 
 ```bash
+# Chat
+
 omniai chat "Who designed the Ruby programming language?"
-```
+omniai chat --provider="google" --model="gemini-2.0-flash" "Who are you?"
 
-```
-The Ruby programming language was created by Yukihiro Matsumoto, often known as "Matz."
-```
+## Speech to Text
 
-```bash
-omniai chat --provider="google" --model="gemini-2.0-flash" "Who are you?"
-```
+omniai speak "Salley sells sea shells by the sea shore." > ./files/audio.wav
 
-```
-I am a large language model, trained by Google.
+# Text to Speech
+
+omniai transcribe "./files/audio.wav"
+
+# Embed
+
+omniai embed "What is the capital of France?"
 ```
 
-### Example #8: [Text-to-Speech](https://github.com/ksylvest/omniai/blob/main/examples/text_to_speech)
+### Example #8: [🔈 Text-to-Speech](https://github.com/ksylvest/omniai/blob/main/examples/text_to_speech)
 
 This example demonstrates using `OmniAI` with **OpenAI** to convert text to speech and save it to a file.
 
@@ -264,7 +268,7 @@ File.open(File.join(__dir__, 'audio.wav'), 'wb') do |file|
 end
 ```
 
-### Example #9: [Speech-to-Text](https://github.com/ksylvest/omniai/blob/main/examples/speech_to_text)
+### Example #9: [🎤 Speech-to-Text](https://github.com/ksylvest/omniai/blob/main/examples/speech_to_text)
 
 This example demonstrates using `OmniAI` with **OpenAI** to convert speech to text.
 
@@ -279,7 +283,7 @@ File.open(File.join(__dir__, 'audio.wav'), 'rb') do |file|
 end
 ```
 
-### Example #10: [Embeddings](https://github.com/ksylvest/omniai/blob/main/examples/embeddings)
+### Example #10: [💻 Embeddings](https://github.com/ksylvest/omniai/blob/main/examples/embeddings)
 
 This example demonstrates using `OmniAI` with **Mistral** to generate embeddings for a dataset. It defines a set of entries (e.g. "George is a teacher." or "Ringo is a doctor.") and then compares the embeddings generated from a query (e.g. "What does George do?" or "Who is a doctor?") to rank the entries by relevance.
 
@@ -322,7 +326,7 @@ search('Who do you call to fix a toilet?')
 'Who do you call to fix a toilet?': 'Paul is a plumber.'
 ```
 
-## Installation
+## 📦 Installation
 
 The main `omniai` gem is installed with:
 
@@ -340,7 +344,7 @@ gem install omniai-google
 gem install omniai-openai
 ```
 
-## Usage
+## 📖 Usage
 
 OmniAI implements APIs for a number of popular clients by default. A client can be initialized using the specific gem (e.g. `omniai-openai` for `OmniAI::OpenAI`). Vendor specific docs can be found within each repo.
 
@@ -362,6 +366,14 @@ require 'omniai/deepseek'
 client = OmniAI::DeepSeek::Client.new
 ```
 
+#### [OmniAI::Llama](https://github.com/ksylvest/omniai-llama)
+
+```ruby
+require 'omniai/llama'
+
+client = OmniAI::Llama::Client.new
+```
+
 #### [OmniAI::Google](https://github.com/ksylvest/omniai-google)
 
 ```ruby
@@ -442,7 +454,7 @@ client = OmniAI::OpenAI::Client.new(timeout: {
 })
 ```
 
-### Chat
+### 💬 Chat
 
 Clients that support chat (e.g. Anthropic w/ "Claude", Google w/ "Gemini", Mistral w/ "LeChat", OpenAI w/ "ChatGPT", etc) generate completions using the following calls:
 
@@ -520,7 +532,7 @@ end
 client.chat('What is the weather in "London" in Celsius and "Paris" in Fahrenheit?', tools: [WeatherTool.new])
 ```
 
-### Transcribe
+### 🎤 Speech to Text
 
 Clients that support transcribe (e.g. OpenAI w/ "Whisper") convert recordings to text via the following calls:
 
@@ -540,7 +552,7 @@ File.open("example.ogg", "rb") do |file|
 end
 ```
 
-### Speak
+### 🔈 Text to Speech
 
 Clients that support speak (e.g. OpenAI w/ "Whisper") convert text to speech via the following calls:
 
@@ -562,7 +574,7 @@ tempfile.close
 tempfile.unlink
 ```
 
-### Embeddings
+### 💻 Embeddings
 
 Clients that support generating embeddings (e.g. OpenAI, Mistral, etc.) convert text to embeddings via the following:
 
@@ -585,7 +597,7 @@ response.embeddings.each do |embedding|
 end
 ```
 
-## CLI
+## 🐚 CLI
 
 OmniAI packages a basic command line interface (CLI) to allow for exploration of various APIs. CLI documentation is available with the `--help` flag:
 
@@ -650,7 +662,19 @@ The capital of Spain is **Madrid**.
 ...
 ```
 
-### MCP
+#### Text-to-Speech
+
+```bash
+omniai speak "Sally sells sea shells on the sea shore." > audio.aac
+```
+
+#### Speech-to-Text
+
+```bash
+omniai transcribe ./audio.aac
+```
+
+## MCP
 
 [MCP](https://modelcontextprotocol.io/introduction) is an open protocol designed to standardize giving context to LLMs. The OmniAI implementation supports building an MCP server that operates via the [stdio](https://modelcontextprotocol.io/docs/concepts/transports) transport.
 

data/lib/omniai/cli/base_handler.rb

@@ -31,7 +31,7 @@ module OmniAI
 
       # @return [OmniAI::Client]
       def client
-        @client ||= OmniAI::Client.find(provider: @provider)
+        @client ||= OmniAI.client(provider: @provider)
       end
     end
   end

data/lib/omniai/cli/speak_handler.rb

@@ -0,0 +1,59 @@
+# frozen_string_literal: true
+
+module OmniAI
+  class CLI
+    # Used by CLI to process commands like:
+    #
+    #    omniai speak "Sally sells seashells by the seashore." > ./audio.aac
+    class SpeakHandler < BaseHandler
+      # @param argv [Array<String>]
+      def handle!(argv:)
+        parser.parse!(argv)
+
+        speak(argv.join(" "))
+      end
+
+    private
+
+      # @param input [String]
+      def speak(input)
+        client.speak(input, **@args) do |chunk|
+          @stdout.write(chunk)
+        end
+        @stdout.flush
+      end
+
+      # @return [OptionParser]
+      def parser
+        OptionParser.new do |options|
+          options.banner = 'usage: omniai speak [options] "<prompt>"'
+
+          options.on("-h", "--help", "help") do
+            @stdout.puts(options)
+            exit
+          end
+
+          options.on("-p", "--provider=PROVIDER", "provider") do |provider|
+            @provider = provider
+          end
+
+          options.on("-m", "--model=MODEL", "model") do |model|
+            @args[:model] = model
+          end
+
+          options.on("-v", "--voice=VOICE", "voice") do |voice|
+            @args[:voice] = voice
+          end
+
+          options.on("-s", "--speed=SPEED", Float, "speed") do |speed|
+            @args[:speed] = speed
+          end
+
+          options.on("-f", "--format=FORMAT", "format") do |format|
+            @args[:format] = format
+          end
+        end
+      end
+    end
+  end
+end

data/lib/omniai/cli/transcribe_handler.rb

@@ -0,0 +1,56 @@
+# frozen_string_literal: true
+
+module OmniAI
+  class CLI
+    # Used by CLI to process commands like:
+    #
+    #    omniai transcribe ./audio.wav
+    class TranscribeHandler < BaseHandler
+      # @param argv [Array<String>]
+      def handle!(argv:)
+        parser.parse!(argv)
+
+        argv.each do |path|
+          transcribe(path)
+        end
+      end
+
+    private
+
+      # @param input [String]
+      def transcribe(input)
+        File.open(input, "rb") do |file|
+          @stdout.puts(client.transcribe(file, **@args).text)
+        end
+      end
+
+      # @return [OptionParser]
+      def parser
+        OptionParser.new do |options|
+          options.banner = 'usage: omniai chat [options] "<prompt>"'
+
+          options.on("-h", "--help", "help") do
+            @stdout.puts(options)
+            exit
+          end
+
+          options.on("-p", "--provider=PROVIDER", "provider") do |provider|
+            @provider = provider
+          end
+
+          options.on("-m", "--model=MODEL", "model") do |model|
+            @args[:model] = model
+          end
+
+          options.on("-l", "--language=LANGUAGE", "language") do |language|
+            @args[:language] = language
+          end
+
+          options.on("-f", "--format=FORMAT", "format") do |format|
+            @args[:format] = format
+          end
+        end
+      end
+    end
+  end
+end

data/lib/omniai/cli.rb

@@ -13,7 +13,7 @@ module OmniAI
     # @param stdin [IO] a stream
     # @param stdout [IO] a stream
     # @param provider [String] a provider
-    def initialize(stdin: $stdin, stdout: $stdout, provider: "openai")
+    def initialize(stdin: $stdin, stdout: $stdout, provider: nil)
       @stdin = stdin
       @stdout = stdout
       @provider = provider
@@ -30,6 +30,8 @@ module OmniAI
         case command
         when "chat" then ChatHandler
         when "embed" then EmbedHandler
+        when "speak" then SpeakHandler
+        when "transcribe" then TranscribeHandler
         else raise Error, "unsupported command=#{command.inspect}"
         end
 
@@ -60,6 +62,9 @@ module OmniAI
         options.separator <<~COMMANDS
           commands:
             chat
+            embed
+            speak
+            transcribe
         COMMANDS
       end
     end

data/lib/omniai/client.rb

@@ -56,6 +56,18 @@ module OmniAI
       raise LoadError, "requires 'omniai-deepseek': `gem install omniai-deepseek`"
     end
 
+    # Lookup the `OmniAI::LLama::Client``. This method requires the provider if it is undefined.
+    #
+    # @raise [OmniAI::LoadError] if the provider is not defined and the gem is not installed
+    #
+    # @return [Class<OmniAI::Client>]
+    def self.llama
+      require "omniai/llama" unless defined?(OmniAI::Llama::Client)
+      OmniAI::Llama::Client
+    rescue ::LoadError
+      raise LoadError, "requires 'omniai-llama': `gem install omniai-llama`"
+    end
+
     # Lookup the `OmniAI::Google::Client``. This method requires the provider if it is undefined.
     #
     # @raise [OmniAI::LoadError] if the provider is not defined and the gem is not installed
@@ -105,11 +117,12 @@ module OmniAI
       end
 
       raise LoadError, <<~TEXT
-        requires 'omniai-openai' or 'omniai-anthropic' or 'openai-deepseek' or 'omniai-google' or 'omniai-mistral':
+        Please run one of the following commands to install a provider specific gem:
 
           `gem install omniai-openai`
           `gem install omniai-anthropic`
           `gem install omniai-deepseek`
+          `gem install omniai-llama`
           `gem install omniai-google`
           `gem install omniai-mistral`
       TEXT
@@ -118,7 +131,7 @@ module OmniAI
     # Initialize a client by provider (e.g. 'openai'). This method attempts to require the provider.
     #
     #
-    # @param provider [String, Symbol] required (e.g. 'anthropic', 'deepseek', 'google', 'mistral', 'openai', etc)
+    # @param provider [String, Symbol] required (e.g. 'anthropic', 'deepseek', 'google', 'mistral', 'openai', 'llama')
     #
     # @raise [OmniAI::LoadError] if the provider is not defined and the gem is not installed
     #
@@ -129,6 +142,7 @@ module OmniAI
         when :anthropic, "anthropic" then anthropic
         when :deepseek, "deepseek" then deepseek
         when :google, "google" then google
+        when :llama, "llama" then llama
         when :mistral, "mistral" then mistral
         when :openai, "openai" then openai
         else raise Error, "unknown provider=#{provider.inspect}"

data/lib/omniai/schema/object.rb

@@ -64,6 +64,16 @@ module OmniAI
         @required = required
       end
 
+      # @return [OmniAI::Schema::Object]
+      def dup
+        self.class.new(
+          title: @title,
+          description: @description,
+          properties: @properties.dup,
+          required: @required.dup
+        )
+      end
+
       # @return [Hash]
       def serialize
         {

data/lib/omniai/speak.rb

@@ -23,7 +23,11 @@ module OmniAI
   #     end
   #   end
   #
-  #   client.transcribe(File.open("..."), model: "...", format: :json)
+  #   File.open('audio.wav', 'wb') do |file|
+  #     client.speak('Sally sells seashells by the seashore.', format: OmniAI::Speak::Format::WAV) do |chunk|
+  #       file << chunk
+  #     end
+  #   end
   class Speak
     module Format
       AAC = "aac"
@@ -34,6 +38,8 @@ module OmniAI
       WAV = "wav"
     end
 
+    DEFAULT_FORMAT = Format::AAC
+
     # @raise [HTTPError]
     #
     # @param client [OmniAI::Client] required
@@ -52,7 +58,7 @@ module OmniAI
     # @yield [chunk]
     #
     # @return [Tempfile]
-    def self.process!(input, client:, model:, voice:, speed: nil, format: nil, &)
+    def self.process!(input, client:, model:, voice:, speed: nil, format: DEFAULT_FORMAT, &)
       new(input, client:, model:, voice:, speed:, format:).process!(&)
     end
 
@@ -68,7 +74,7 @@ module OmniAI
     #   - "opus"
     #   - "pcm"
     #   - "wav"
-    def initialize(input, client:, model:, voice:, speed: nil, format: nil)
+    def initialize(input, client:, model:, voice:, speed: nil, format: DEFAULT_FORMAT)
       @input = input
       @client = client
       @model = model

data/lib/omniai/tool.rb

@@ -16,18 +16,20 @@ module OmniAI
   #   end
   class Tool
     class << self
-      # @param description [String]
+      # @param description [String] optional
       #
       # @return [String]
       def description(description = nil)
-        return @description if description.nil?
-
-        @description = description
+        if description.nil?
+          @description || (superclass.description unless superclass.eql?(OmniAI::Tool))
+        else
+          @description = description
+        end
       end
 
       # @return [OmniAI::Schema::Object]
       def parameters
-        @parameters ||= OmniAI::Schema::Object.new
+        @parameters ||= superclass.eql?(OmniAI::Tool) ? superclass.parameters.dup : OmniAI::Schema::Object.new
       end
 
       # @param name [Symbol]
@@ -38,7 +40,7 @@ module OmniAI
 
       # @param names [Array<Symbol>]
       def required(names)
-        parameters.required = names
+        parameters.required += names
       end
 
       # Converts a class name to a tool:

data/lib/omniai/transcribe.rb

@@ -23,7 +23,9 @@ module OmniAI
   #     end
   #   end
   #
-  #   client.transcribe(File.open("..."), model: "...", format: :json)
+  #   File.open('audio.wav', 'rb') do |file|
+  #     client.transcribe(file, model: "...", format: :json)
+  #   end
   class Transcribe
     module Language
       AFRIKAANS = "af"

data/lib/omniai/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module OmniAI
-  VERSION = "2.7.0"
+  VERSION = "2.8.1"
 end

data/lib/omniai.rb

@@ -25,9 +25,9 @@ loader.setup
 module OmniAI
   class Error < StandardError; end
 
-  # Discover a client by provider ('openai' then 'anthropic' then 'google' then 'mistral' then 'deepseek').
+  # Discover a client by provider ('openai' then 'anthropic' then 'google' then 'mistral' then 'deepseek' then 'llama').
   #
-  # @param provider [Symbol] the provider to use (e.g. :openai, :anthropic, :google, :mistral, :deepseek) - optional
+  # @param provider [Symbol] the provider to use (e.g. :openai, :anthropic, :google, :mistral, :deepseek, :llama)
   #
   # @raise [OmniAI::LoadError] if no providers are installed
   #

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification
 name: omniai
 version: !ruby/object:Gem::Version
-  version: 2.7.0
+  version: 2.8.1
 platform: ruby
 authors:
 - Kevin Sylvestre
 bindir: exe
 cert_chain: []
-date: 2025-05-22 00:00:00.000000000 Z
+date: 2025-05-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: event_stream_parser
@@ -103,6 +103,8 @@ files:
 - lib/omniai/cli/base_handler.rb
 - lib/omniai/cli/chat_handler.rb
 - lib/omniai/cli/embed_handler.rb
+- lib/omniai/cli/speak_handler.rb
+- lib/omniai/cli/transcribe_handler.rb
 - lib/omniai/client.rb
 - lib/omniai/config.rb
 - lib/omniai/context.rb