cohere-ruby 0.9.10 → 1.0.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 6c5f0625529c315bd0a953bb8e0acf36fbebe685b13decde1fe26d93b5d97ab8
-  data.tar.gz: 504ce83377f6f2175b2d7ab9f899c307a7a02563c75b8c630b249d0e7487c746
+  metadata.gz: 0c65ac65e6fa623709455907a71697f721311d2933e96a10a5c49f907f356f02
+  data.tar.gz: 553fd85c32a2a99b2a8d1a75f1a2aa0a0c9abec290f305ed26caa855515f717b
 SHA512:
-  metadata.gz: 1dc8c2c85e551438f597346679c3bb198808e906eb9317dd05c26511556e34f512707cbc4bb3046487de9e498ca7b92c6405d2f9c3e4175f466e49618c7215b9
-  data.tar.gz: 0024a4c8c6091fa78fe83dcd7879ea8a511436e4ddc16929dd3e88e517901ba1ac569e16c519f47abccbba08a0cdd43fe795c4e867b4e252f0d5003a9ffb5b3e
+  metadata.gz: bb7dd10864d789f91225f9370011e622dc6353a55a7039b8ef509610980c1503c3916f210189361cc1dcc20cb1db880d7fd4d80f967867fdcecd6803e85e8bdb
+  data.tar.gz: 89eb351c192abfda2c905810210b21be5d6a5b18aa6e3d5a4ac28efaba9fc23e1a0525510b51a4399c5e7086fab12a854035388806c4401122c3463277941990

data/.env.example

@@ -0,0 +1 @@
+COHERE_API_KEY=

data/CHANGELOG.md

@@ -1,5 +1,13 @@
 ## [Unreleased]
 
+## [1.0.1] - 2024-11-22
+
+## [1.0.0] - 2024-11-22
+- Migrate to v2 APIs
+
+## [0.9.11] - 2024-08-01
+- New `rerank()` method
+
 ## [0.9.10] - 2024-05-10
 - /chat endpoint does not require `message:` parameter anymore
 

data/Gemfile

@@ -9,3 +9,4 @@ gem "rake", "~> 13.0"
 
 gem "rspec", "~> 3.0"
 gem "standard", "~> 1.28.0"
+gem "dotenv", "~> 2.8.1"

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    cohere-ruby (0.9.10)
+    cohere-ruby (1.0.1)
       faraday (>= 2.0.1, < 3.0)
 
 GEM
@@ -9,6 +9,7 @@ GEM
   specs:
     ast (2.4.2)
     diff-lcs (1.5.0)
+    dotenv (2.8.1)
     faraday (2.7.10)
       faraday-net_http (>= 2.0, < 3.1)
       ruby2_keywords (>= 0.0.4)
@@ -74,6 +75,7 @@ PLATFORMS
 
 DEPENDENCIES
   cohere-ruby!
+  dotenv (~> 2.8.1)
   rake (~> 13.0)
   rspec (~> 3.0)
   standard (~> 1.28.0)

data/README.md

@@ -1,19 +1,19 @@
 # Cohere
 
 <p>
-    <img alt='Weaviate logo' src='https://static.wikia.nocookie.net/logopedia/images/d/d4/Cohere_2023.svg/revision/latest?cb=20230419182227' height='50' />
+    <img alt='Cohere logo' src='https://static.wikia.nocookie.net/logopedia/images/d/d4/Cohere_2023.svg/revision/latest?cb=20230419182227' height='50' />
     +&nbsp;&nbsp;
     <img alt='Ruby logo' src='https://user-images.githubusercontent.com/541665/230231593-43861278-4550-421d-a543-fd3553aac4f6.png' height='40' />
 </p>
 
 Cohere API client for Ruby.
 
-Part of the [Langchain.rb](https://github.com/andreibondarev/langchainrb) stack.
+Part of the [Langchain.rb](https://github.com/patterns-ai-core/langchainrb) stack.
 
-![Tests status](https://github.com/andreibondarev/cohere-ruby/actions/workflows/ci.yml/badge.svg)
+![Tests status](https://github.com/patterns-ai-core/cohere-ruby/actions/workflows/ci.yml/badge.svg)
 [![Gem Version](https://badge.fury.io/rb/cohere-ruby.svg)](https://badge.fury.io/rb/cohere-ruby)
 [![Docs](http://img.shields.io/badge/yard-docs-blue.svg)](http://rubydoc.info/gems/cohere-ruby)
-[![License](https://img.shields.io/badge/license-MIT-green.svg)](https://github.com/andreibondarev/cohere-ruby/blob/main/LICENSE.txt)
+[![License](https://img.shields.io/badge/license-MIT-green.svg)](https://github.com/patterns-ai-core/cohere-ruby/blob/main/LICENSE.txt)
 [![](https://dcbadge.vercel.app/api/server/WDARp7J2n8?compact=true&style=flat)](https://discord.gg/WDARp7J2n8)
 
 ## Installation
@@ -42,7 +42,7 @@ client = Cohere::Client.new(
 
 ```ruby
 client.generate(
-    prompt: "Once upon a time in a magical land called"
+  prompt: "Once upon a time in a magical land called"
 )
 ```
 
@@ -50,14 +50,18 @@ client.generate(
 
 ```ruby
 client.chat(
-    message: "Hey! How are you?"
+  model: "command-r-plus-08-2024",
+  messages: [{role:"user", content: "Hey! How are you?"}]
 )
 ```
 
 `chat` supports a streaming option. You can pass a block to the `chat` method and it will yield a new chunk as soon as it is received.
 
 ```ruby
-client.chat(message: "Hey! How are you?", stream: true) do |chunk, overall_received_bytes|
+client.chat(
+  model: "command-r-plus-08-2024",
+  messages: [{role:"user", content: "Hey! How are you?"}]
+) do |chunk, overall_received_bytes|
   puts "Received #{overall_received_bytes} bytes: #{chunk.force_encoding(Encoding::UTF_8)}"
 end
 ```
@@ -68,35 +72,54 @@ end
 
 ```ruby
 tools = [
-   {
-     name: "query_daily_sales_report",
-     description: "Connects to a database to retrieve overall sales volumes and sales information for a given day.",
-     parameter_definitions: {
-       day: {
-         description: "Retrieves sales data for this day, formatted as YYYY-MM-DD.",
-         type: "str",
-         required: true
-       }
-     }
-   }
+  {
+    name: "query_daily_sales_report",
+    description: "Connects to a database to retrieve overall sales volumes and sales information for a given day.",
+    parameter_definitions: {
+      day: {
+        description: "Retrieves sales data for this day, formatted as YYYY-MM-DD.",
+        type: "str",
+        required: true
+      }
+    }
+  }
 ]
 
 message = "Can you provide a sales summary for 29th September 2023, and also give me some details about the products in the 'Electronics' category, for example their prices and stock levels?"
 
 client.chat(
   model: model,
-  message: message,
-  tools: tools,
+  messages: [{ role:"user", content: message }],
+  tools: tools
 )
 ```
 
-
-
 ### Embed
 
 ```ruby
 client.embed(
-    texts: ["hello!"]
+  model: "embed-english-v3.0",
+  texts: ["hello", "goodbye"],
+  input_type: "classification",
+  embedding_types: ["float"]
+)
+```
+
+### Rerank
+
+```ruby
+docs = [
+  "Carson City is the capital city of the American state of Nevada.",
+  "The Commonwealth of the Northern Mariana Islands is a group of islands in the Pacific Ocean. Its capital is Saipan.",
+  "Capitalization or capitalisation in English grammar is the use of a capital letter at the start of a word. English usage varies from capitalization in other languages.",
+  "Washington, D.C. (also known as simply Washington or D.C., and officially as the District of Columbia) is the capital of the United States. It is a federal district.",
+  "Capital punishment (the death penalty) has existed in the United States since beforethe United States was a country. As of 2017, capital punishment is legal in 30 of the 50 states.",
+]
+
+client.rerank(
+  model: "rerank-english-v3.0",
+  query: "What is the capital of the United States?",
+  documents: docs
 )
 ```
 
@@ -104,16 +127,16 @@ client.embed(
 
 ```ruby
 examples = [
-    { text: "Dermatologists don't like her!", label: "Spam" },
-    { text: "Hello, open to this?", label: "Spam" },
-    { text: "I need help please wire me $1000 right now", label: "Spam" },
-    { text: "Nice to know you ;)", label: "Spam" },
-    { text: "Please help me?", label: "Spam" },
-    { text: "Your parcel will be delivered today", label: "Not spam" },
-    { text: "Review changes to our Terms and Conditions", label: "Not spam" },
-    { text: "Weekly sync notes", label: "Not spam" },
-    { text: "Re: Follow up from today's meeting", label: "Not spam" },
-    { text: "Pre-read for tomorrow", label: "Not spam" }
+  { text: "Dermatologists don't like her!", label: "Spam" },
+  { text: "Hello, open to this?", label: "Spam" },
+  { text: "I need help please wire me $1000 right now", label: "Spam" },
+  { text: "Nice to know you ;)", label: "Spam" },
+  { text: "Please help me?", label: "Spam" },
+  { text: "Your parcel will be delivered today", label: "Not spam" },
+  { text: "Review changes to our Terms and Conditions", label: "Not spam" },
+  { text: "Weekly sync notes", label: "Not spam" },
+  { text: "Re: Follow up from today's meeting", label: "Not spam" },
+  { text: "Pre-read for tomorrow", label: "Not spam" }
 ]
 
 inputs = [
@@ -122,8 +145,9 @@ inputs = [
 ]
 
 client.classify(
-    examples: examples,
-    inputs: inputs
+  model: "embed-multilingual-v2.0",
+  inputs: inputs,
+  examples: examples
 )
 ```
 
@@ -131,7 +155,8 @@ client.classify(
 
 ```ruby
 client.tokenize(
-    text: "hello world!"
+  model: "command-r-plus-08-2024",
+  text: "Hello, world!"
 )
 ```
 
@@ -139,7 +164,8 @@ client.tokenize(
 
 ```ruby
 client.detokenize(
-    tokens: [33555, 1114 , 34]
+  model: "command-r-plus-08-2024",
+  tokens: [33555, 1114, 34]
 )
 ```
 
@@ -147,7 +173,7 @@ client.detokenize(
 
 ```ruby
 client.detect_language(
-    texts: ["Здравствуй, Мир"]
+  texts: ["Здравствуй, Мир"]
 )
 ```
 
@@ -155,7 +181,7 @@ client.detect_language(
 
 ```ruby
 client.summarize(
-    text: "..."
+  text: "..."
 )
 ```
 

data/lib/cohere/client.rb

@@ -6,62 +6,56 @@ module Cohere
   class Client
     attr_reader :api_key, :connection
 
-    ENDPOINT_URL = "https://api.cohere.ai/v1"
-
     def initialize(api_key:, timeout: nil)
       @api_key = api_key
       @timeout = timeout
     end
 
+    # Generates a text response to a user message and streams it down, token by token
     def chat(
-      message: nil,
-      model: nil,
+      model:,
+      messages:,
       stream: false,
-      preamble: nil,
-      preamble_override: nil,
-      chat_history: [],
-      conversation_id: nil,
-      prompt_truncation: nil,
-      connectors: [],
-      search_queries_only: false,
+      tools: [],
       documents: [],
-      citation_quality: nil,
-      temperature: nil,
+      citation_options: nil,
+      response_format: nil,
+      safety_mode: nil,
       max_tokens: nil,
-      k: nil,
-      p: nil,
+      stop_sequences: nil,
+      temperature: nil,
       seed: nil,
       frequency_penalty: nil,
       presence_penalty: nil,
-      tools: [],
+      k: nil,
+      p: nil,
+      logprops: nil,
       &block
     )
-      response = connection.post("chat") do |req|
+      response = v2_connection.post("chat") do |req|
         req.body = {}
 
-        req.body[:message] = message if message
-        req.body[:model] = model if model
-        if stream || block
-          req.body[:stream] = true
-          req.options.on_data = block if block
-        end
-        req.body[:preamble] = preamble if preamble
-        req.body[:preamble_override] = preamble_override if preamble_override
-        req.body[:chat_history] = chat_history if chat_history
-        req.body[:conversation_id] = conversation_id if conversation_id
-        req.body[:prompt_truncation] = prompt_truncation if prompt_truncation
-        req.body[:connectors] = connectors if connectors
-        req.body[:search_queries_only] = search_queries_only if search_queries_only
-        req.body[:documents] = documents if documents
-        req.body[:citation_quality] = citation_quality if citation_quality
-        req.body[:temperature] = temperature if temperature
+        req.body[:model] = model
+        req.body[:messages] = messages if messages
+        req.body[:tools] = tools if tools.any?
+        req.body[:documents] = documents if documents.any?
+        req.body[:citation_options] = citation_options if citation_options
+        req.body[:response_format] = response_format if response_format
+        req.body[:safety_mode] = safety_mode if safety_mode
         req.body[:max_tokens] = max_tokens if max_tokens
-        req.body[:k] = k if k
-        req.body[:p] = p if p
+        req.body[:stop_sequences] = stop_sequences if stop_sequences
+        req.body[:temperature] = temperature if temperature
         req.body[:seed] = seed if seed
         req.body[:frequency_penalty] = frequency_penalty if frequency_penalty
         req.body[:presence_penalty] = presence_penalty if presence_penalty
-        req.body[:tools] = tools if tools
+        req.body[:k] = k if k
+        req.body[:p] = p if p
+        req.body[:logprops] = logprops if logprops
+
+        if stream || block
+          req.body[:stream] = true
+          req.options.on_data = block if block
+        end
       end
       response.body
     end
@@ -84,7 +78,7 @@ module Cohere
       logit_bias: nil,
       truncate: nil
     )
-      response = connection.post("generate") do |req|
+      response = v1_connection.post("generate") do |req|
         req.body = {prompt: prompt}
         req.body[:model] = model if model
         req.body[:num_generations] = num_generations if num_generations
@@ -104,56 +98,90 @@ module Cohere
       response.body
     end
 
+    # This endpoint returns text embeddings. An embedding is a list of floating point numbers that captures semantic information about the text that it represents.
     def embed(
-      texts:,
-      model: nil,
-      input_type: nil,
+      model:,
+      input_type:,
+      embedding_types:,
+      texts: nil,
+      images: nil,
       truncate: nil
     )
-      response = connection.post("embed") do |req|
-        req.body = {texts: texts}
-        req.body[:model] = model if model
-        req.body[:input_type] = input_type if input_type
+      response = v2_connection.post("embed") do |req|
+        req.body = {
+          model: model,
+          input_type: input_type,
+          embedding_types: embedding_types
+        }
+        req.body[:texts] = texts if texts
+        req.body[:images] = images if images
         req.body[:truncate] = truncate if truncate
       end
       response.body
     end
 
+    # This endpoint takes in a query and a list of texts and produces an ordered array with each text assigned a relevance score.
+    def rerank(
+      model:,
+      query:,
+      documents:,
+      top_n: nil,
+      rank_fields: nil,
+      return_documents: nil,
+      max_chunks_per_doc: nil
+    )
+      response = v2_connection.post("rerank") do |req|
+        req.body = {
+          model: model,
+          query: query,
+          documents: documents
+        }
+        req.body[:top_n] = top_n if top_n
+        req.body[:rank_fields] = rank_fields if rank_fields
+        req.body[:return_documents] = return_documents if return_documents
+        req.body[:max_chunks_per_doc] = max_chunks_per_doc if max_chunks_per_doc
+      end
+      response.body
+    end
+
+    # This endpoint makes a prediction about which label fits the specified text inputs best.
     def classify(
+      model:,
       inputs:,
-      examples:,
-      model: nil,
-      present: nil,
+      examples: nil,
+      preset: nil,
       truncate: nil
     )
-      response = connection.post("classify") do |req|
+      response = v1_connection.post("classify") do |req|
         req.body = {
-          inputs: inputs,
-          examples: examples
+          model: model,
+          inputs: inputs
         }
-        req.body[:model] = model if model
-        req.body[:present] = present if present
+        req.body[:examples] = examples if examples
+        req.body[:preset] = preset if preset
         req.body[:truncate] = truncate if truncate
       end
       response.body
     end
 
-    def tokenize(text:, model: nil)
-      response = connection.post("tokenize") do |req|
-        req.body = model.nil? ? {text: text} : {text: text, model: model}
+    # This endpoint splits input text into smaller units called tokens using byte-pair encoding (BPE).
+    def tokenize(text:, model:)
+      response = v1_connection.post("tokenize") do |req|
+        req.body = {text: text, model: model}
       end
       response.body
     end
 
-    def detokenize(tokens:, model: nil)
-      response = connection.post("detokenize") do |req|
-        req.body = model.nil? ? {tokens: tokens} : {tokens: tokens, model: model}
+    # This endpoint takes tokens using byte-pair encoding and returns their text representation.
+    def detokenize(tokens:, model:)
+      response = v1_connection.post("detokenize") do |req|
+        req.body = {tokens: tokens, model: model}
       end
       response.body
     end
 
     def detect_language(texts:)
-      response = connection.post("detect-language") do |req|
+      response = v1_connection.post("detect-language") do |req|
         req.body = {texts: texts}
       end
       response.body
@@ -168,7 +196,7 @@ module Cohere
       temperature: nil,
       additional_command: nil
     )
-      response = connection.post("summarize") do |req|
+      response = v1_connection.post("summarize") do |req|
         req.body = {text: text}
         req.body[:length] = length if length
         req.body[:format] = format if format
@@ -182,17 +210,22 @@ module Cohere
 
     private
 
-    # standard:disable Lint/DuplicateMethods
-    def connection
-      @connection ||= Faraday.new(url: ENDPOINT_URL, request: {timeout: @timeout}) do |faraday|
-        if api_key
-          faraday.request :authorization, :Bearer, api_key
-        end
+    def v1_connection
+      @connection ||= Faraday.new(url: "https://api.cohere.ai/v1", request: {timeout: @timeout}) do |faraday|
+        faraday.request :authorization, :Bearer, api_key
+        faraday.request :json
+        faraday.response :json, content_type: /\bjson$/
+        faraday.adapter Faraday.default_adapter
+      end
+    end
+
+    def v2_connection
+      @connection ||= Faraday.new(url: "https://api.cohere.com/v2", request: {timeout: @timeout}) do |faraday|
+        faraday.request :authorization, :Bearer, api_key
         faraday.request :json
         faraday.response :json, content_type: /\bjson$/
         faraday.adapter Faraday.default_adapter
       end
     end
-    # standard:enable Lint/DuplicateMethods
   end
 end

data/lib/cohere/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Cohere
-  VERSION = "0.9.10"
+  VERSION = "1.0.1"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: cohere-ruby
 version: !ruby/object:Gem::Version
-  version: 0.9.10
+  version: 1.0.1
 platform: ruby
 authors:
 - Andrei Bondarev
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2024-05-11 00:00:00.000000000 Z
+date: 2024-11-23 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: faraday
@@ -37,6 +37,7 @@ executables: []
 extensions: []
 extra_rdoc_files: []
 files:
+- ".env.example"
 - ".rspec"
 - CHANGELOG.md
 - Gemfile
@@ -70,7 +71,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.1
+rubygems_version: 3.5.11
 signing_key:
 specification_version: 4
 summary: Cohere API client for Ruby.