llm.rb 0.7.0 → 0.7.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: a7175b2fe81c74e007dd41db2e0fe1bd3f3639bed375af25da0f8ed2778ea2b5
-  data.tar.gz: 1c752e61cb288fed412b342b66279e7dfdb0337705e33af3e2a1deb1d408b8d0
+  metadata.gz: 7d5d93a645b666da3d6947c2076189063aec26e7bc3381cfb84d6a6aea4ce8fa
+  data.tar.gz: db764cd8e9180a3c21ca5bf2b35d8a5fa3f525cb63101381aca09f9e92cb5d37
 SHA512:
-  metadata.gz: 9af91ba96e63b2c43c7f6a836db5fed48da19ba1f2bdbb48894cc71fb940eca261930fec6b8fd9a6f641fc9c69402de9cdd6fd7f9cad9a7035b69ddad04de65a
-  data.tar.gz: b3f8af44ebb2522aba58621d19a19424805471f6f9a6f8ec834ebe417e6aa08e9f0233ef19a12695e829d44ea61b164ae57697379aebdf24450b829fdc04ac25
+  metadata.gz: c17999419e02e8c2d9d689299d149dd76077b1f52154557253dd9fe3876ff1109eaaa29d7d7aceda8147f38aa572261b455eaa253fbede212b874af42c8e03e0
+  data.tar.gz: 1061af3f752a1e8cbc37c8929a23f6bfa97328f66f3adf030e435f468785ffa6a3aae32fe5120936f872b60d8a48a31fcbf35cee0de66b8ee12ddedf17228beb

data/README.md

@@ -22,7 +22,7 @@ images, files, and JSON Schema generation.
 - 🗣️ Text-to-speech, transcription, and translation
 - 🖼️ Image generation, editing, and variation support
 - 📎 File uploads and prompt-aware file interaction
-- 💡 Multimodal prompts (text, URLs, files)
+- 💡 Multimodal prompts (text, images, PDFs, URLs, files)
 
 #### Embeddings
 - 🧮 Text embeddings and vector support
@@ -78,8 +78,8 @@ llm = LLM.voyageai(key: "yourapikey")
 
 The following example enables lazy mode for a
 [LLM::Chat](https://0x1eef.github.io/x/llm.rb/LLM/Chat.html)
-object by entering into a "lazy" conversation where messages are buffered and
-sent to the provider only when necessary.  Both lazy and non-lazy conversations
+object by entering into a conversation where messages are buffered and
+sent to the provider only when necessary. Both lazy and non-lazy conversations
 maintain a message thread that can be reused as context throughout a conversation.
 The example captures the spirit of llm.rb by demonstrating how objects cooperate
 together through composition, and it uses the stateless chat completions API that
@@ -97,6 +97,8 @@ msgs = bot.chat do |prompt|
   prompt.user "Tell me the answer to (5 + 15) * 2"
   prompt.user "Tell me the answer to ((5 + 15) * 2) / 10"
 end
+
+# At this point, we execute a single request
 msgs.each { print "[#{_1.role}] ", _1.content, "\n" }
 
 ##
@@ -134,18 +136,18 @@ require "llm"
 
 llm = LLM.openai(key: ENV["KEY"])
 schema = llm.schema.object({fruit: llm.schema.string.enum("Apple", "Orange", "Pineapple")})
-bot = LLM::Chat.new(llm, schema:)
+bot = LLM::Chat.new(llm, schema:).lazy
 bot.chat "Your favorite fruit is Pineapple", role: :system
 bot.chat "What fruit is your favorite?", role: :user
 bot.messages.find(&:assistant?).content! # => {fruit: "Pineapple"}
 
 schema = llm.schema.object({answer: llm.schema.integer.required})
-bot = LLM::Chat.new(llm, schema:)
+bot = LLM::Chat.new(llm, schema:).lazy
 bot.chat "Tell me the answer to ((5 + 5) / 2)", role: :user
 bot.messages.find(&:assistant?).content! # => {answer: 5}
 
 schema = llm.schema.object({probability: llm.schema.number.required})
-bot = LLM::Chat.new(llm, schema:)
+bot = LLM::Chat.new(llm, schema:).lazy
 bot.chat "Does the earth orbit the sun?", role: :user
 bot.messages.find(&:assistant?).content! # => {probability: 1}
 ```
@@ -175,14 +177,18 @@ arbitrary commands from a LLM without sanitizing the input first :) Without furt
 #!/usr/bin/env ruby
 require "llm"
 
-llm = LLM.openai(key: ENV["KEY"])
+llm  = LLM.openai(key: ENV["KEY"])
 tool = LLM.function(:system) do |fn|
   fn.description "Run a shell command"
   fn.params do |schema|
     schema.object(command: schema.string.required)
   end
   fn.define do |params|
-    system(params.command)
+    ro, wo = IO.pipe
+    re, we = IO.pipe
+    Process.wait Process.spawn(params.command, out: wo, err: we)
+    [wo,we].each(&:close)
+    {stderr: re.read, stdout: ro.read}
   end
 end
 
@@ -196,8 +202,8 @@ bot.chat "What operating system am I running? (short version please!)", role: :u
 bot.chat bot.functions.map(&:call) # report return value to the LLM
 
 ##
-# Thu May  1 10:01:02 UTC 2025
-# FreeBSD
+# {stderr: "", stdout: "Thu May  1 10:01:02 UTC 2025"}
+# {stderr: "", stdout: "FreeBSD"}
 ```
 
 ### Audio

data/lib/llm/chat/builder.rb

@@ -6,6 +6,10 @@ class LLM::Chat
   module Builder
     private
 
+    ##
+    # @param [String] prompt The prompt
+    # @param [Hash] params
+    # @return [LLM::Response::Respond]
     def create_response!(prompt, params)
       @provider.responses.create(
         prompt,
@@ -13,6 +17,10 @@ class LLM::Chat
       )
     end
 
+    ##
+    # @param [String] prompt The prompt
+    # @param [Hash] params
+    # @return [LLM::Response::Completion]
     def create_completion!(prompt, params)
       @provider.complete(
         prompt,

data/lib/llm/chat/conversable.rb

@@ -6,22 +6,42 @@ class LLM::Chat
   module Conversable
     private
 
+    ##
+    # Queues a response to be sent to the provider.
+    # @param [String] prompt The prompt
+    # @param [Hash] params
+    # @return [void]
     def async_response(prompt, params = {})
       role = params.delete(:role)
       @messages << [LLM::Message.new(role, prompt), @params.merge(params), :respond]
     end
 
+    ##
+    # Sends a response to the provider and returns the response.
+    # @param [String] prompt The prompt
+    # @param [Hash] params
+    # @return [LLM::Response::Respond]
     def sync_response(prompt, params = {})
       role = params[:role]
       @response = create_response!(prompt, params)
       @messages.concat [Message.new(role, prompt), @response.outputs[0]]
     end
 
+    ##
+    # Queues a completion to be sent to the provider.
+    # @param [String] prompt The prompt
+    # @param [Hash] params
+    # @return [void]
     def async_completion(prompt, params = {})
       role = params.delete(:role)
       @messages.push [LLM::Message.new(role, prompt), @params.merge(params), :complete]
     end
 
+    ##
+    # Sends a completion to the provider and returns the completion.
+    # @param [String] prompt The prompt
+    # @param [Hash] params
+    # @return [LLM::Response::Completion]
     def sync_completion(prompt, params = {})
       role = params[:role]
       completion = create_completion!(prompt, params)

data/lib/llm/multipart.rb

@@ -27,20 +27,6 @@ class LLM::Multipart
     "multipart/form-data; boundary=#{@boundary}"
   end
 
-  ##
-  # Returns the multipart request body parts
-  # @return [Array<String>]
-  def parts
-    params.map do |key, value|
-      locals = {key: key.to_s.b, boundary: boundary.to_s.b}
-      if value.respond_to?(:path)
-        file_part(key, value, locals)
-      else
-        data_part(key, value, locals)
-      end
-    end
-  end
-
   ##
   # Returns the multipart request body
   # @return [String]
@@ -54,47 +40,63 @@ class LLM::Multipart
 
   attr_reader :params
 
-  def attributes(file)
-    {
-      filename: File.basename(file.path).b,
-      content_type: LLM::Mime[file].b
-    }
+  def file(locals, file)
+    locals = locals.merge(attributes(file))
+    build_file(locals) do |body|
+      IO.copy_stream(file.path, body)
+      body << "\r\n"
+    end
   end
 
-  def multipart_header(type:, locals:)
-    if type == :file
-      str = StringIO.new("".b)
-      str << "--#{locals[:boundary]}" \
+  def form(locals, value)
+    locals = locals.merge(value:)
+    build_form(locals) do |body|
+      body << value.to_s
+      body << "\r\n"
+    end
+  end
+
+  def build_file(locals)
+    StringIO.new("".b).tap do |io|
+      io << "--#{locals[:boundary]}" \
              "\r\n" \
              "Content-Disposition: form-data; name=\"#{locals[:key]}\";" \
              "filename=\"#{locals[:filename]}\"" \
              "\r\n" \
              "Content-Type: #{locals[:content_type]}" \
              "\r\n\r\n"
-    elsif type == :data
-      str = StringIO.new("".b)
-      str << "--#{locals[:boundary]}" \
+      yield(io)
+    end
+  end
+
+  def build_form(locals)
+    StringIO.new("".b).tap do |io|
+      io << "--#{locals[:boundary]}" \
              "\r\n" \
              "Content-Disposition: form-data; name=\"#{locals[:key]}\"" \
              "\r\n\r\n"
-    else
-      raise "unknown type: #{type}"
+      yield(io)
     end
   end
 
-  def file_part(key, file, locals)
-    locals = locals.merge(attributes(file))
-    multipart_header(type: :file, locals:).tap do |io|
-      IO.copy_stream(file.path, io)
-      io << "\r\n"
+  ##
+  # Returns the multipart request body parts
+  # @return [Array<String>]
+  def parts
+    params.map do |key, value|
+      locals = {key: key.to_s.b, boundary: boundary.to_s.b}
+      if value.respond_to?(:path)
+        file(locals, value)
+      else
+        form(locals, value)
+      end
     end
   end
 
-  def data_part(key, value, locals)
-    locals = locals.merge(value:)
-    multipart_header(type: :data, locals:).tap do |io|
-      io << value.to_s
-      io << "\r\n"
-    end
+  def attributes(file)
+    {
+      filename: File.basename(file.path).b,
+      content_type: LLM::Mime[file].b
+    }
   end
 end

data/lib/llm/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module LLM
-  VERSION = "0.7.0"
+  VERSION = "0.7.1"
 end

data/llm.gemspec

@@ -8,14 +8,15 @@ Gem::Specification.new do |spec|
   spec.authors = ["Antar Azri", "0x1eef"]
   spec.email = ["azantar@proton.me", "0x1eef@proton.me"]
 
-  spec.summary = "llm.rb is a lightweight Ruby library that provides a " \
-                 "common interface and set of functionality for multple " \
-                 "Large Language Models (LLMs). It is designed to be simple, " \
-                 "flexible, and easy to use."
+  spec.summary = "llm.rb is a zero-dependency Ruby toolkit for " \
+                 "Large Language Models that includes OpenAI, Gemini, " \
+                 "Anthropic, Ollama, and LlamaCpp. It’s fast, simple " \
+                 "and composable – with full support for chat, tool calling, audio, " \
+                 "images, files, and JSON Schema generation."
   spec.description = spec.summary
   spec.homepage = "https://github.com/llmrb/llm"
   spec.license = "0BSDL"
-  spec.required_ruby_version = ">= 3.0.0"
+  spec.required_ruby_version = ">= 3.2.0"
 
   spec.metadata["homepage_uri"] = spec.homepage
   spec.metadata["source_code_uri"] = "https://github.com/llmrb/llm"

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: llm.rb
 version: !ruby/object:Gem::Version
-  version: 0.7.0
+  version: 0.7.1
 platform: ruby
 authors:
 - Antar Azri
@@ -9,7 +9,7 @@ authors:
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2025-05-09 00:00:00.000000000 Z
+date: 2025-05-11 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: webmock
@@ -151,9 +151,10 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: '2.8'
-description: llm.rb is a lightweight Ruby library that provides a common interface
-  and set of functionality for multple Large Language Models (LLMs). It is designed
-  to be simple, flexible, and easy to use.
+description: llm.rb is a zero-dependency Ruby toolkit for Large Language Models that
+  includes OpenAI, Gemini, Anthropic, Ollama, and LlamaCpp. It’s fast, simple and
+  composable – with full support for chat, tool calling, audio, images, files, and
+  JSON Schema generation.
 email:
 - azantar@proton.me
 - 0x1eef@proton.me
@@ -259,7 +260,7 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 3.0.0
+      version: 3.2.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
@@ -269,7 +270,8 @@ requirements: []
 rubygems_version: 3.5.23
 signing_key:
 specification_version: 4
-summary: llm.rb is a lightweight Ruby library that provides a common interface and
-  set of functionality for multple Large Language Models (LLMs). It is designed to
-  be simple, flexible, and easy to use.
+summary: llm.rb is a zero-dependency Ruby toolkit for Large Language Models that includes
+  OpenAI, Gemini, Anthropic, Ollama, and LlamaCpp. It’s fast, simple and composable
+  – with full support for chat, tool calling, audio, images, files, and JSON Schema
+  generation.
 test_files: []