cloudflare-ai 0.8.0 → 0.9.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: dfe55c49c1310aa3940e9bc8963535680a93ccf90a0b98061b26a75c85fb42d3
-  data.tar.gz: 79335c8d797ca3585f8fb7a770fd064ac3aa7e76622fd8b794fbbb88abf24b43
+  metadata.gz: c727f28c32fdfa8aace3c440ccd7c4f99e0edf9c60a35605dd386dfb1e83f3d9
+  data.tar.gz: 2a405251e064c93fc9de0bb4f27da57e0bdb342400cfc867ae4a92384494af4b
 SHA512:
-  metadata.gz: ee82ccc27724c218043e4702a1dd39ba0878c6ff35caa35592e9dae3b26ea7bd1ccb0b73b182ad97864d637414edcfb1a587ec4b203fc4fe30767f15d56f4d74
-  data.tar.gz: 4cab183f89a4fd13cca87e8b0e264c93eddae4bf12c62a7ccfb5ae9e453b4cabc6916a3fdbf573ca55291ee8c537df190ebae06dbb2e2377af208e10cca8fe0c
+  metadata.gz: f10b2c9ee7dbdfef2ac61c8a425bfb6d1f06ccddab9296c86cfcf097f243e499b5498dbcd0163ac03e31c5ae845f9ab54bcd506f73e2533ce172eb360336dfd4
+  data.tar.gz: a01a5743d7d845813582317b50d9bc131c981c884191eff0890286d2a85881801f9eda768ec6246693192aa086a5fd771580e2fcab4865f881e054bc810cdcaf

data/lib/cloudflare/ai/client.rb

@@ -12,6 +12,13 @@ class Cloudflare::AI::Client
     @api_token = api_token
   end
 
+  def caption(image: nil, model_name: Cloudflare::AI::Models.image_to_text.first)
+    url = service_url_for(account_id: account_id, model_name: model_name)
+
+    image = File.open(image) if image.is_a?(String)
+    Cloudflare::AI::Results::ImageToText.new(post_request_with_binary_file(url, image).body)
+  end
+
   def chat(messages:, model_name: default_text_generation_model_name, max_tokens: default_max_tokens, &block)
     url = service_url_for(account_id: account_id, model_name: model_name)
     stream = block ? true : false
@@ -41,6 +48,13 @@ class Cloudflare::AI::Client
     post_streamable_request(url, payload, &block)
   end
 
+  def detect_objects(image: nil, model_name: Cloudflare::AI::Models.object_detection.first)
+    url = service_url_for(account_id: account_id, model_name: model_name)
+
+    image = File.open(image) if image.is_a?(String)
+    Cloudflare::AI::Results::ObjectDetection.new(post_request_with_binary_file(url, image).body)
+  end
+
   def draw(prompt:, num_steps: 20, model_name: Cloudflare::AI::Models.text_to_image.first)
     url = service_url_for(account_id: account_id, model_name: model_name)
     payload = {prompt: prompt, num_steps: num_steps}.to_json
@@ -59,6 +73,13 @@ class Cloudflare::AI::Client
     Cloudflare::AI::Results::TextEmbedding.new(connection.post(url, payload).body)
   end
 
+  def summarize(text:, model_name: Cloudflare::AI::Models.summarization.first, max_tokens: 1024)
+    url = service_url_for(account_id: account_id, model_name: model_name)
+    payload = {input_text: text, max_tokens: max_tokens}.to_json
+
+    Cloudflare::AI::Results::Summarization.new(connection.post(url, payload).body)
+  end
+
   def transcribe(source_url: nil, audio: nil, model_name: Cloudflare::AI::Models.automatic_speech_recognition.first)
     raise ArgumentError, "Must provide either audio_url or audio" if [source_url, audio].compact.size != 1
 

data/lib/cloudflare/ai/models.rb

@@ -1,43 +1,90 @@
 class Cloudflare::AI::Models
   class << self
-    def text_generation
-      %w[@cf/meta/llama-2-7b-chat-fp16 @cf/meta/llama-2-7b-chat-int8 @cf/mistral/mistral-7b-instruct-v0.1 @hf/thebloke/codellama-7b-instruct-awq]
+    def all
+      {
+        automatic_speech_recognition: automatic_speech_recognition,
+        image_classification: image_classification,
+        image_to_text: image_to_text,
+        object_detection: object_detection,
+        summarization: summarization,
+        text_classification: text_classification,
+        text_embeddings: text_embedding,
+        text_generation: text_generation,
+        text_to_image: text_to_image,
+        translation: translation
+      }
     end
 
     def automatic_speech_recognition
       %w[@cf/openai/whisper]
     end
 
-    def translation
-      %w[@cf/meta/m2m100-1.2b]
+    def image_classification
+      %w[@cf/microsoft/resnet-50]
     end
 
-    def text_classification
-      %w[@cf/huggingface/distilbert-sst-2-int8]
+    def image_to_text
+      %w[@cf/unum/uform-gen2-qwen-500m]
     end
 
-    def image_classification
-      %w[@cf/microsoft/resnet-50]
+    def object_detection
+      %w[@cf/meta/det-resnet-50]
     end
 
-    def text_to_image
-      %w[@cf/stabilityai/stable-diffusion-xl-base-1.0]
+    def summarization
+      %w[@cf/facebook/bart-large-cnn]
+    end
+
+    def text_classification
+      %w[@cf/huggingface/distilbert-sst-2-int8]
     end
 
     def text_embedding
       %w[@cf/baai/bge-base-en-v1.5 @cf/baai/bge-large-en-v1.5 @cf/baai/bge-small-en-v1.5]
     end
 
-    def all
-      {
-        text_generation: text_generation,
-        automatic_speech_recognition: automatic_speech_recognition,
-        translation: translation,
-        text_classification: text_classification,
-        image_classification: image_classification,
-        text_to_image: text_to_image,
-        text_embeddings: text_embedding
-      }
+    def text_generation
+      %w[
+        @hf/thebloke/codellama-7b-instruct-awq
+        @hf/thebloke/deepseek-coder-6.7b-base-awq
+        @hf/thebloke/deepseek-coder-6.7b-instruct-awq
+        @cf/deepseek-ai/deepseek-math-7b-base
+        @cf/deepseek-ai/deepseek-math-7b-instruct
+        @cf/thebloke/discolm-german-7b-v1-awq
+        @cf/tiiuae/falcon-7b-instruct
+        @hf/thebloke/llama-2-13b-chat-awq
+        @cf/meta/llama-2-7b-chat-fp16
+        @cf/meta/llama-2-7b-chat-int8
+        @hf/thebloke/llamaguard-7b-awq
+        @cf/mistral/mistral-7b-instruct-v0.1
+        @hf/thebloke/mistral-7b-instruct-v0.1-awq
+        @hf/thebloke/neural-chat-7b-v3-1-awq
+        @hf/thebloke/openchat_3.5-awq
+        @cf/openchat/openchat-3.5-0106
+        @hf/thebloke/openhermes-2.5-mistral-7b-awq
+        @cf/microsoft/phi-2
+        @cf/qwen/qwen1.5-0.5b-chat
+        @cf/qwen/qwen1.5-1.8b-chat
+        @cf/qwen/qwen1.5-14b-chat-awq
+        @cf/qwen/qwen1.5-7b-chat-awq
+        @cf/defog/sqlcoder-7b-2
+        @cf/tinyllama/tinyllama-1.1b-chat-v1.0
+        @hf/thebloke/zephyr-7b-beta-awq
+      ]
+    end
+
+    def text_to_image
+      %w[
+        @cf/lykon/dreamshaper-8-lcm
+        @cf/runwayml/stable-diffusion-v1-5-img2img
+        @cf/runwayml/stable-diffusion-v1-5-inpainting
+        @cf/stabilityai/stable-diffusion-xl-base-1.0
+        @cf/bytedance/stable-diffusion-xl-lightning
+      ]
+    end
+
+    def translation
+      %w[@cf/meta/m2m100-1.2b]
     end
   end
 end

data/lib/cloudflare/ai/results/image_to_text.rb

@@ -0,0 +1,3 @@
+class Cloudflare::AI::Results::ImageToText < Cloudflare::AI::Result
+  # Empty seam kept for consistency with other result objects that have more complexity.
+end

data/lib/cloudflare/ai/results/object_detection.rb

@@ -0,0 +1,3 @@
+class Cloudflare::AI::Results::ObjectDetection < Cloudflare::AI::Result
+  # Empty seam kept for consistency with other result objects that have more complexity.
+end

data/lib/cloudflare/ai/results/summarization.rb

@@ -0,0 +1,5 @@
+class Cloudflare::AI::Results::Summarization < Cloudflare::AI::Result
+  def summary
+    result&.dig(:summary) # nil if no response
+  end
+end

data/lib/cloudflare/ai/version.rb

@@ -2,6 +2,6 @@
 
 module Cloudflare
   module AI
-    VERSION = "0.8.0"
+    VERSION = "0.9.1"
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: cloudflare-ai
 version: !ruby/object:Gem::Version
-  version: 0.8.0
+  version: 0.9.1
 platform: ruby
 authors:
 - Ajay Krishnan
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2024-01-26 00:00:00.000000000 Z
+date: 2024-02-28 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: activemodel
@@ -115,6 +115,9 @@ files:
 - lib/cloudflare/ai/result.rb
 - lib/cloudflare/ai/results/automatic_speech_recognition.rb
 - lib/cloudflare/ai/results/image_classification.rb
+- lib/cloudflare/ai/results/image_to_text.rb
+- lib/cloudflare/ai/results/object_detection.rb
+- lib/cloudflare/ai/results/summarization.rb
 - lib/cloudflare/ai/results/text_classification.rb
 - lib/cloudflare/ai/results/text_embedding.rb
 - lib/cloudflare/ai/results/text_generation.rb