[Workers AI] Add pricing and model maturity (#13226)

* first draft * added pricing calculator image * added pricing tables * few copy tweaks * Update models * Add conditional beta pill to model page title * Make non-beta models float to top of model index * Add weight to sort models * Add redirects * Changes from PR review * Replace pricing calc image * Apply suggestions from code review * Apply suggestions from code review K -> ,000 * Fix wrangler hyperdrive binding key (#13190) * [R2] runtime API: TypeScript formatting audit (#13200) * [R2] runtime API: TypeScript formatting audit * more edits * [SSL] Fix mTLS for Workers redirect after bindings reorg (#13211) * Fix mtls for workers redirect after bidings reorg * Remove 301 to add again * Re-add 301 to push and retrigger build * connect your domain (#12701) * first draft * updating quotes * additional updates * Apply suggestions from code review Co-authored-by: abstergail <[email protected]> * updates with some feedback * updates * removing product filter, so fundamentals glossary is for all terms * updating index pages for better mobile experience * updates * updating redirects * fixing link * updating dns limitations on the proxy status page. updating how cloudflare works feedback * moving what is cloudflare changes to another branch * restoring original file * merging an older commit in * updates * updating weights and links * revisions * Apply suggestions from code review Co-authored-by: abstergail <[email protected]> * couple more updates * final edits * Apply suggestions from code review Co-authored-by: marciocloudflare <[email protected]> * fixing meta titles --------- Co-authored-by: abstergail <[email protected]> Co-authored-by: marciocloudflare <[email protected]> * Revert Workers external_link pages as redirect targets (#13214) * [WAI] codeowner update (#13216) * [WAI] codeowner update * npm run build --------- Co-authored-by: Kody Jackson <[email protected]> * Update smart-placement.md (#12455) * Minor corrections (#12995) "Developer" -> "Develop" "Workers" -> "Worker's" * [D1] Update Wrangler commands (#12336) * [D1] Update Wrangler commands * Apply suggestions from code review Co-authored-by: Kate Tungusova <[email protected]> * Remove npx from d1 commands * Apply suggestions from code review --------- Co-authored-by: Kate Tungusova <[email protected]> * Document the workers no-importscripts compat flag (#12618) * Document the workers no-importscripts compat flag Refs: cloudflare/workerd#1556 * Update content/workers/_partials/_platform-compatibility-dates/no-importscripts.md Co-authored-by: Kate Tungusova <[email protected]> --------- Co-authored-by: Kate Tungusova <[email protected]> * Clarify subrequest billing (#12873) We don't bill for subrequests, but old language left it a bit ambiguous. This simplifies. * [Gateway] Update app types (#12948) Co-authored-by: Kate Tungusova <[email protected]> * Clarify pricing <> Service Bindings (#11669) * Link to Workers Trace Events documentation from Trace Events Logpush (#12307) Currently this page doesn't actually explain what fields are available in logs, if you use Workers Trace Events Logpush. This fixes by adding a link to this page. * resolving comments --------- Co-authored-by: Michelle Chen <[email protected]> Co-authored-by: Kate Tungusova <[email protected]> Co-authored-by: Kane Wang <[email protected]> Co-authored-by: Rebecca Tamachiro <[email protected]> Co-authored-by: jason-cf <[email protected]> Co-authored-by: abstergail <[email protected]> Co-authored-by: marciocloudflare <[email protected]> Co-authored-by: Kody Jackson <[email protected]> Co-authored-by: Matt Silverlock <[email protected]> Co-authored-by: Jürg Gutknecht <[email protected]> Co-authored-by: Kian <[email protected]> Co-authored-by: James M Snell <[email protected]> Co-authored-by: Brendan Irvine-Broque <[email protected]> Co-authored-by: Max Phillips <[email protected]> Co-authored-by: MC <[email protected]>
cloudflare · Mar 1, 2024 · 3a01a53 · 3a01a53
1 parent 1ddf81c
commit 3a01a53
Show file tree

Hide file tree

Showing 49 changed files with 421 additions and 2,870 deletions.
diff --git a/assets/images/workers-ai/pricing-calculator.png b/assets/images/workers-ai/pricing-calculator.png
diff --git a/content/_redirects b/content/_redirects
@@ -1265,6 +1265,9 @@
 /workers-ai/models/translation/  /workers-ai/models/#translation 301
 /workers-ai/models/text-classification/ /workers-ai/models/#text-classification 301
 /workers-ai/models/image-classification/ /workers-ai/models/#image-classification 301
+/workers-ai/models/codellama-7b-instruct-awq/ /workers-ai/models/ 301
+/workers-ai/models/openchat_3.5-awq/ /workers-ai/models/ 301
+
 
 # workers KV
 /kv/platform/environments/ /kv/reference/environments/ 301

diff --git a/content/workers-ai/_index.md b/content/workers-ai/_index.md
@@ -15,32 +15,34 @@ meta:
 Run machine learning models, powered by serverless GPUs, on Cloudflare's global network.
 {{</description>}}
 
+{{<render file="_billing-announcement.md">}}
+
 Workers AI allows you to run machine learning models, on the Cloudflare network, from your own code -- whether that be from Workers, Pages, or anywhere via REST API.
 
 ### Designed for developers
 
 You shouldn't have to understand the ins and outs of machine learning to leverage the power of it. With a focus on developer experience, it's powerful, yet simple, and only a few lines of code to get started:
 
 ```js
-import { Ai } from '@cloudflare/ai';
+import { Ai } from "@cloudflare/ai";
 
-const ai = new Ai(env.AI)
+const ai = new Ai(env.AI);
 
-const output = await ai.run('@cf/meta/llama-2-7b-chat-int8', {
-  prompt: 'Tell me about Workers AI'
-})
+const output = await ai.run("@cf/meta/llama-2-7b-chat-int8", {
+  prompt: "Tell me about Workers AI",
+});
 ```
 
 ### Models you know and love
 
-Workers AI comes with a curated set of popular open-source models that *just work*.
+Workers AI comes with a curated set of popular open-source models that _just work_.
 
 Supporting multiple classes of models, here are some AI tasks you can unlock with Workers AI:
 
-* **Natural language processing** - text generation + summarization + classification + translation, similarity analysis, question answering
-* **Image generation** - text to image generation tasks
-* **Computer Vision** - image classification, object detection
-* **Audio** - Automatic speech recognition (ASR)
+- **Natural language processing** - text generation + summarization + classification + translation, similarity analysis, question answering
+- **Image generation** - text to image generation tasks
+- **Computer Vision** - image classification, object detection
+- **Audio** - Automatic speech recognition (ASR)
 
 ### Runs on a global network of GPUs
 

diff --git a/content/workers-ai/_partials/_billing-announcement.md b/content/workers-ai/_partials/_billing-announcement.md
@@ -0,0 +1,12 @@
+---
+_build:
+  publishResources: false
+  render: never
+  list: never
+---
+
+{{<Aside type="note">}}
+
+Workers AI will begin billing for usage on non-beta models after April 1, 2024. Refer to [Pricing](/workers-ai/platform/pricing) for more details.
+
+{{</Aside>}}
diff --git a/content/workers-ai/models/_index.md b/content/workers-ai/models/_index.md
@@ -9,4 +9,6 @@ hideChildren: true
 
 Browse our entire catalog of models.
 
-{{% models-grouped-by-task %}}
+{{<render file="_billing-announcement.md">}}
+
+{{% models-grouped-by-task %}}
diff --git a/content/workers-ai/models/bart-large-cnn.md b/content/workers-ai/models/bart-large-cnn.md
@@ -10,10 +10,13 @@ model:
     description: "Summarization is the task of producing a shorter version of a document while preserving its important information. Some models can extract text from the original input, while other models can generate entirely new text."
   tags:
     - "summarization"
-  properties: []
+  properties:
+    - property_id: "beta"
+      value: "true"
 task_type: "summarization"
 model_display_name: "bart-large-cnn"
 layout: "model"
+weight: 0
 title: "bart-large-cnn"
 json_schema:
   input: "{\n  \"type\": \"object\",\n  \"properties\": {\n    \"input_text\": {\n      \"type\": \"string\"\n    },\n    \"max_length\": {\n      \"type\": \"integer\",\n      \"default\": 1024\n    }\n  },\n  \"required\": [\n    \"input_text\"\n  ]\n}"

diff --git a/content/workers-ai/models/bge-base-en-v1.5.md b/content/workers-ai/models/bge-base-en-v1.5.md
@@ -20,9 +20,12 @@ model:
       value: "768"
     - property_id: "constellation_config"
       value: "infer_response_cache: in_memory\n\nmax_requests_per_min:\n  default: 180\n  accounts:\n    32118455: 2160 # ai.cloudflare.com staging\n    50147400: 2160 # ai.cloudflare.com\n\nneurons:\n  metrics:\n    - name: input_tokens\n      neuron_cost: 0.0060575\nmax_concurrent_requests: 100"
+    - property_id: "beta"
+      value: "false"
 task_type: "text-embeddings"
 model_display_name: "bge-base-en-v1.5"
 layout: "model"
+weight: 100
 title: "bge-base-en-v1.5"
 json_schema:
   input: "{\n  \"type\": \"object\",\n  \"properties\": {\n    \"text\": {\n      \"oneOf\": [\n        {\n          \"type\": \"string\"\n        },\n        {\n          \"type\": \"array\",\n          \"items\": {\n            \"type\": \"string\"\n          },\n          \"maxItems\": 100\n        }\n      ]\n    }\n  },\n  \"required\": [\n    \"text\"\n  ]\n}"

diff --git a/content/workers-ai/models/bge-large-en-v1.5.md b/content/workers-ai/models/bge-large-en-v1.5.md
@@ -20,9 +20,12 @@ model:
       value: "1024"
     - property_id: "constellation_config"
       value: "infer_response_cache: in_memory\n\nmax_requests_per_min:\n  default: 120\n\nneurons:\n  metrics:\n    - name: input_tokens\n      neuron_cost: 0.01858166667\nmax_concurrent_requests: 100"
+    - property_id: "beta"
+      value: "false"
 task_type: "text-embeddings"
 model_display_name: "bge-large-en-v1.5"
 layout: "model"
+weight: 100
 title: "bge-large-en-v1.5"
 json_schema:
   input: "{\n  \"type\": \"object\",\n  \"properties\": {\n    \"text\": {\n      \"oneOf\": [\n        {\n          \"type\": \"string\"\n        },\n        {\n          \"type\": \"array\",\n          \"items\": {\n            \"type\": \"string\"\n          },\n          \"maxItems\": 100\n        }\n      ]\n    }\n  },\n  \"required\": [\n    \"text\"\n  ]\n}"

diff --git a/content/workers-ai/models/bge-small-en-v1.5.md b/content/workers-ai/models/bge-small-en-v1.5.md
@@ -20,9 +20,12 @@ model:
       value: "384"
     - property_id: "constellation_config"
       value: "infer_response_cache: in_memory\n\nmax_requests_per_min:\n  default: 180\n\nneurons:\n  metrics:\n    - name: input_tokens\n      neuron_cost: 0.001840833333\nmax_concurrent_requests: 100"
+    - property_id: "beta"
+      value: "false"
 task_type: "text-embeddings"
 model_display_name: "bge-small-en-v1.5"
 layout: "model"
+weight: 100
 title: "bge-small-en-v1.5"
 json_schema:
   input: "{\n  \"type\": \"object\",\n  \"properties\": {\n    \"text\": {\n      \"oneOf\": [\n        {\n          \"type\": \"string\"\n        },\n        {\n          \"type\": \"array\",\n          \"items\": {\n            \"type\": \"string\"\n          },\n          \"maxItems\": 100\n        }\n      ]\n    }\n  },\n  \"required\": [\n    \"text\"\n  ]\n}"

diff --git a/content/workers-ai/models/codellama-7b-instruct-awq.md b/content/workers-ai/models/codellama-7b-instruct-awq.md
diff --git a/content/workers-ai/models/deepseek-coder-6.7b-base-awq.md b/content/workers-ai/models/deepseek-coder-6.7b-base-awq.md
@@ -14,9 +14,12 @@ model:
   properties:
     - property_id: "terms"
       value: "https://huggingface.co/TheBloke/deepseek-coder-6.7B-base-AWQ"
+    - property_id: "beta"
+      value: "true"
 task_type: "text-generation"
 model_display_name: "deepseek-coder-6.7b-base-awq"
 layout: "model"
+weight: 0
 title: "deepseek-coder-6.7b-base-awq"
 json_schema:
   input: "{\n  \"type\": \"object\",\n  \"oneOf\": [\n    {\n      \"properties\": {\n        \"prompt\": {\n          \"type\": \"string\",\n          \"maxLength\": 4096\n        },\n        \"raw\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"stream\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"max_tokens\": {\n          \"type\": \"integer\",\n          \"default\": 256\n        }\n      },\n      \"required\": [\n        \"prompt\"\n      ]\n    },\n    {\n      \"properties\": {\n        \"messages\": {\n          \"type\": \"array\",\n          \"items\": {\n            \"type\": \"object\",\n            \"properties\": {\n              \"role\": {\n                \"type\": \"string\"\n              },\n              \"content\": {\n                \"type\": \"string\",\n                \"maxLength\": 4096\n              }\n            },\n            \"required\": [\n              \"role\",\n              \"content\"\n            ]\n          }\n        },\n        \"stream\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"max_tokens\": {\n          \"type\": \"integer\",\n          \"default\": 256\n        }\n      },\n      \"required\": [\n        \"messages\"\n      ]\n    }\n  ]\n}"

diff --git a/content/workers-ai/models/deepseek-coder-6.7b-instruct-awq.md b/content/workers-ai/models/deepseek-coder-6.7b-instruct-awq.md
@@ -14,9 +14,12 @@ model:
   properties:
     - property_id: "terms"
       value: "https://huggingface.co/TheBloke/deepseek-coder-6.7B-instruct-AWQ"
+    - property_id: "beta"
+      value: "true"
 task_type: "text-generation"
 model_display_name: "deepseek-coder-6.7b-instruct-awq"
 layout: "model"
+weight: 0
 title: "deepseek-coder-6.7b-instruct-awq"
 json_schema:
   input: "{\n  \"type\": \"object\",\n  \"oneOf\": [\n    {\n      \"properties\": {\n        \"prompt\": {\n          \"type\": \"string\",\n          \"maxLength\": 4096\n        },\n        \"raw\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"stream\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"max_tokens\": {\n          \"type\": \"integer\",\n          \"default\": 256\n        }\n      },\n      \"required\": [\n        \"prompt\"\n      ]\n    },\n    {\n      \"properties\": {\n        \"messages\": {\n          \"type\": \"array\",\n          \"items\": {\n            \"type\": \"object\",\n            \"properties\": {\n              \"role\": {\n                \"type\": \"string\"\n              },\n              \"content\": {\n                \"type\": \"string\",\n                \"maxLength\": 4096\n              }\n            },\n            \"required\": [\n              \"role\",\n              \"content\"\n            ]\n          }\n        },\n        \"stream\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"max_tokens\": {\n          \"type\": \"integer\",\n          \"default\": 256\n        }\n      },\n      \"required\": [\n        \"messages\"\n      ]\n    }\n  ]\n}"

diff --git a/content/workers-ai/models/deepseek-math-7b-base.md b/content/workers-ai/models/deepseek-math-7b-base.md
@@ -10,10 +10,13 @@ model:
     description: "Family of generative text models, such as large language models (LLM), that can be adapted for a variety of natural language tasks."
   tags:
     - "text-generation"
-  properties: []
+  properties:
+    - property_id: "beta"
+      value: "true"
 task_type: "text-generation"
 model_display_name: "deepseek-math-7b-base"
 layout: "model"
+weight: 0
 title: "deepseek-math-7b-base"
 json_schema:
   input: "{\n  \"type\": \"object\",\n  \"oneOf\": [\n    {\n      \"properties\": {\n        \"prompt\": {\n          \"type\": \"string\",\n          \"maxLength\": 4096\n        },\n        \"raw\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"stream\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"max_tokens\": {\n          \"type\": \"integer\",\n          \"default\": 256\n        }\n      },\n      \"required\": [\n        \"prompt\"\n      ]\n    },\n    {\n      \"properties\": {\n        \"messages\": {\n          \"type\": \"array\",\n          \"items\": {\n            \"type\": \"object\",\n            \"properties\": {\n              \"role\": {\n                \"type\": \"string\"\n              },\n              \"content\": {\n                \"type\": \"string\",\n                \"maxLength\": 4096\n              }\n            },\n            \"required\": [\n              \"role\",\n              \"content\"\n            ]\n          }\n        },\n        \"stream\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"max_tokens\": {\n          \"type\": \"integer\",\n          \"default\": 256\n        }\n      },\n      \"required\": [\n        \"messages\"\n      ]\n    }\n  ]\n}"

diff --git a/content/workers-ai/models/deepseek-math-7b-instruct.md b/content/workers-ai/models/deepseek-math-7b-instruct.md
@@ -15,9 +15,12 @@ model:
       value: "https://huggingface.co/deepseek-ai/deepseek-math-7b-instruct"
     - property_id: "terms"
       value: "https://github.com/deepseek-ai/DeepSeek-Math/blob/main/LICENSE-MODEL"
+    - property_id: "beta"
+      value: "true"
 task_type: "text-generation"
 model_display_name: "deepseek-math-7b-instruct"
 layout: "model"
+weight: 0
 title: "deepseek-math-7b-instruct"
 json_schema:
   input: "{\n  \"type\": \"object\",\n  \"oneOf\": [\n    {\n      \"properties\": {\n        \"prompt\": {\n          \"type\": \"string\",\n          \"maxLength\": 4096\n        },\n        \"raw\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"stream\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"max_tokens\": {\n          \"type\": \"integer\",\n          \"default\": 256\n        }\n      },\n      \"required\": [\n        \"prompt\"\n      ]\n    },\n    {\n      \"properties\": {\n        \"messages\": {\n          \"type\": \"array\",\n          \"items\": {\n            \"type\": \"object\",\n            \"properties\": {\n              \"role\": {\n                \"type\": \"string\"\n              },\n              \"content\": {\n                \"type\": \"string\",\n                \"maxLength\": 4096\n              }\n            },\n            \"required\": [\n              \"role\",\n              \"content\"\n            ]\n          }\n        },\n        \"stream\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"max_tokens\": {\n          \"type\": \"integer\",\n          \"default\": 256\n        }\n      },\n      \"required\": [\n        \"messages\"\n      ]\n    }\n  ]\n}"

diff --git a/content/workers-ai/models/detr-resnet-50.md b/content/workers-ai/models/detr-resnet-50.md
@@ -1,25 +1,22 @@
 ---
 model:
-  id: "6ade40e1-1646-4013-8da0-95abb3360e33"
+  id: "cc34ce52-3059-415f-9a48-12aa919d37ee"
   source: 1
-  name: "@cf/meta/detr-resnet-50"
-  description: "DEtection TRansformer (DETR) model with ResNet-50 backbone for image object detection."
+  name: "@cf/facebook/detr-resnet-50"
+  description: "DEtection TRansformer (DETR) model trained end-to-end on COCO 2017 object detection (118k annotated images)."
   task:
     id: "9c178979-90d9-49d8-9e2c-0f1cf01815d4"
     name: "Object Detection"
     description: "Object detection models can detect instances of objects like persons, faces, license plates, or others in an image. This task takes an image as input and returns a list of detected objects, each one containing a label, a probability score, and its surrounding box coordinates."
   tags:
-    - "experimental"
-    - "meta"
-    - "text-generation"
+    - "object-detection"
   properties:
-    - property_id: "info"
-      value: "https://huggingface.co/facebook/detr-resnet-50"
-    - property_id: "constellation_config"
-      value: "max_requests_per_min:\n  default: 180\n\nneurons:\n  cost_per_infer: 0.6841666668  # ...3x that of resnet-50 (:shrug:).\nmax_concurrent_requests: 100"
+    - property_id: "beta"
+      value: "true"
 task_type: "object-detection"
 model_display_name: "detr-resnet-50"
 layout: "model"
+weight: 0
 title: "detr-resnet-50"
 json_schema:
   input: "{\n  \"oneOf\": [\n    {\n      \"type\": \"string\",\n      \"format\": \"binary\"\n    },\n    {\n      \"type\": \"object\",\n      \"properties\": {\n        \"image\": {\n          \"type\": \"array\",\n          \"items\": {\n            \"type\": \"number\"\n          }\n        }\n      }\n    }\n  ]\n}"

diff --git a/content/workers-ai/models/discolm-german-7b-v1-awq.md b/content/workers-ai/models/discolm-german-7b-v1-awq.md
@@ -13,9 +13,12 @@ model:
   properties:
     - property_id: "info"
       value: "https://huggingface.co/TheBloke/DiscoLM_German_7b_v1-AWQ"
+    - property_id: "beta"
+      value: "true"
 task_type: "text-generation"
 model_display_name: "discolm-german-7b-v1-awq"
 layout: "model"
+weight: 0
 title: "discolm-german-7b-v1-awq"
 json_schema:
   input: "{\n  \"type\": \"object\",\n  \"oneOf\": [\n    {\n      \"properties\": {\n        \"prompt\": {\n          \"type\": \"string\",\n          \"maxLength\": 4096\n        },\n        \"raw\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"stream\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"max_tokens\": {\n          \"type\": \"integer\",\n          \"default\": 256\n        }\n      },\n      \"required\": [\n        \"prompt\"\n      ]\n    },\n    {\n      \"properties\": {\n        \"messages\": {\n          \"type\": \"array\",\n          \"items\": {\n            \"type\": \"object\",\n            \"properties\": {\n              \"role\": {\n                \"type\": \"string\"\n              },\n              \"content\": {\n                \"type\": \"string\",\n                \"maxLength\": 4096\n              }\n            },\n            \"required\": [\n              \"role\",\n              \"content\"\n            ]\n          }\n        },\n        \"stream\": {\n          \"type\": \"boolean\",\n          \"default\": false\n        },\n        \"max_tokens\": {\n          \"type\": \"integer\",\n          \"default\": 256\n        }\n      },\n      \"required\": [\n        \"messages\"\n      ]\n    }\n  ]\n}"

diff --git a/content/workers-ai/models/distilbert-sst-2-int8.md b/content/workers-ai/models/distilbert-sst-2-int8.md
@@ -14,9 +14,12 @@ model:
   properties:
     - property_id: "info"
       value: "https://huggingface.co/Intel/distilbert-base-uncased-finetuned-sst-2-english-int8-static"
+    - property_id: "beta"
+      value: "false"
 task_type: "text-classification"
 model_display_name: "distilbert-sst-2-int8"
 layout: "model"
+weight: 100
 title: "distilbert-sst-2-int8"
 json_schema:
   input: "{\n  \"type\": \"object\",\n  \"properties\": {\n    \"text\": {\n      \"type\": \"string\"\n    }\n  },\n  \"required\": [\n    \"text\"\n  ]\n}"