bedrock-wrapper 2.2.0 → 2.3.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (12) hide show
  1. package/CHANGELOG.md +17 -0
  2. package/README.md +74 -21
  3. package/bedrock-models.js +507 -419
  4. package/bedrock-wrapper.js +223 -36
  5. package/docs/bedrock-wrapper.jpg +0 -0
  6. package/interactive-example.js +1 -0
  7. package/package.json +42 -38
  8. package/test-image.jpg +0 -0
  9. package/test-models-output.txt +22 -6
  10. package/test-models.js +145 -145
  11. package/test-vision.js +71 -0
  12. package/utils.js +27 -27
package/CHANGELOG.md CHANGED
@@ -1,6 +1,23 @@
1
1
  # Changelog
2
2
  All notable changes to this project will be documented in this file.
3
3
 
4
+ ## [2.3.0] - 2025-02-15 (Claude 3.7 & Image Support)
5
+ ### Added
6
+ - Support for Claude 3.7 models
7
+ - Claude-3-7-Sonnet
8
+ - Claude-3-7-Sonnet-Thinking
9
+ - Image support for compatible Claude models
10
+ - Claude 3.5 Sonnet
11
+ - Claude 3.5 Sonnet v2
12
+ - Claude 3.7 Sonnet
13
+ - Claude 3.7 Sonnet Thinking
14
+ - Enhanced message handling for multimodal content
15
+ - Documentation for image support usage
16
+
17
+ ### Changed
18
+ - Updated model configuration for image-capable models
19
+ - Improved response handling for multimodal inputs
20
+
4
21
  ## [2.2.0] - 2025-01-01 (Llama 3.3 70b)
5
22
  ### Added
6
23
  - Support for Llama 3.3 70b
package/README.md CHANGED
@@ -1,6 +1,9 @@
1
1
  # 🪨 Bedrock Wrapper
2
+
2
3
  Bedrock Wrapper is an npm package that simplifies the integration of existing OpenAI-compatible API objects with AWS Bedrock's serverless inference LLMs. Follow the steps below to integrate into your own application, or alternativly use the 🔀 [Bedrock Proxy Endpoint](https://github.com/jparkerweb/bedrock-proxy-endpoint) project to spin up your own custom OpenAI server endpoint for even easier inference (using the standard `baseUrl`, and `apiKey` params).
3
4
 
5
+ ![bedrock-wrapper](https://raw.githubusercontent.com/jparkerweb/bedrock-wrapper/refs/heads/main/docs/bedrock-wrapper.jpg)
6
+
4
7
  ---
5
8
 
6
9
  ### Maintained by
@@ -62,7 +65,7 @@ Bedrock Wrapper is an npm package that simplifies the integration of existing Op
62
65
  ]
63
66
  ```
64
67
 
65
- ***the `model` value should be either a corresponding `modelName` or `modelId` for the supported `bedrock_models` (see the Supported Models section below)***
68
+ ***the `model` value should be the corresponding `modelName` value in the `bedrock_models` section below (see Supported Models below)***
66
69
 
67
70
  4. call the `bedrockWrapper` function and pass in the previously defined `awsCreds` and `openaiChatCompletionsCreateObject` objects
68
71
  ```javascript
@@ -98,26 +101,28 @@ Bedrock Wrapper is an npm package that simplifies the integration of existing Op
98
101
 
99
102
  ### Supported Models
100
103
 
101
- | modelName | modelId |
102
- |----------------------|-------------------------------------------|
103
- | Claude-3-5-Sonnet-v2 | anthropic.claude-3-5-sonnet-20241022-v2:0 |
104
- | Claude-3-5-Sonnet | anthropic.claude-3-5-sonnet-20240620-v1:0 |
105
- | Claude-3-5-Haiku | anthropic.claude-3-5-haiku-20241022-v1:0 |
106
- | Claude-3-Haiku | anthropic.claude-3-haiku-20240307-v1:0 |
107
- | Llama-3-3-70b | us.meta.llama3-3-70b-instruct-v1:0 |
108
- | Llama-3-2-1b | us.meta.llama3-2-1b-instruct-v1:0 |
109
- | Llama-3-2-3b | us.meta.llama3-2-3b-instruct-v1:0 |
110
- | Llama-3-2-11b | us.meta.llama3-2-11b-instruct-v1:0 |
111
- | Llama-3-2-90b | us.meta.llama3-2-90b-instruct-v1:0 |
112
- | Llama-3-1-8b | meta.llama3-1-8b-instruct-v1:0 |
113
- | Llama-3-1-70b | meta.llama3-1-70b-instruct-v1:0 |
114
- | Llama-3-1-405b | meta.llama3-1-405b-instruct-v1:0 |
115
- | Llama-3-8b | meta.llama3-8b-instruct-v1:0 |
116
- | Llama-3-70b | meta.llama3-70b-instruct-v1:0 |
117
- | Mistral-7b | mistral.mistral-7b-instruct-v0:2 |
118
- | Mixtral-8x7b | mistral.mixtral-8x7b-instruct-v0:1 |
119
- | Mistral-Large | mistral.mistral-large-2402-v1:0 |
120
-
104
+ | modelName | AWS Model Id | Image |
105
+ |----------------------------|----------------------------------------------|-------|
106
+ | Claude-3-7-Sonnet-Thinking | us.anthropic.claude-3-7-sonnet-20250219-v1:0 | ✅ |
107
+ | Claude-3-7-Sonnet | us.anthropic.claude-3-7-sonnet-20250219-v1:0 | ✅ |
108
+ | Claude-3-5-Sonnet-v2 | anthropic.claude-3-5-sonnet-20241022-v2:0 | ✅ |
109
+ | Claude-3-5-Sonnet | anthropic.claude-3-5-sonnet-20240620-v1:0 | ✅ |
110
+ | Claude-3-5-Haiku | anthropic.claude-3-5-haiku-20241022-v1:0 | ❌ |
111
+ | Claude-3-Haiku | anthropic.claude-3-haiku-20240307-v1:0 | ❌ |
112
+ | Llama-3-3-70b | us.meta.llama3-3-70b-instruct-v1:0 | ❌ |
113
+ | Llama-3-2-1b | us.meta.llama3-2-1b-instruct-v1:0 | ❌ |
114
+ | Llama-3-2-3b | us.meta.llama3-2-3b-instruct-v1:0 | ❌ |
115
+ | Llama-3-2-11b | us.meta.llama3-2-11b-instruct-v1:0 | ❌ |
116
+ | Llama-3-2-90b | us.meta.llama3-2-90b-instruct-v1:0 | ❌ |
117
+ | Llama-3-1-8b | meta.llama3-1-8b-instruct-v1:0 | ❌ |
118
+ | Llama-3-1-70b | meta.llama3-1-70b-instruct-v1:0 | ❌ |
119
+ | Llama-3-1-405b | meta.llama3-1-405b-instruct-v1:0 | ❌ |
120
+ | Llama-3-8b | meta.llama3-8b-instruct-v1:0 | ❌ |
121
+ | Llama-3-70b | meta.llama3-70b-instruct-v1:0 | ❌ |
122
+ | Mistral-7b | mistral.mistral-7b-instruct-v0:2 | ❌ |
123
+ | Mixtral-8x7b | mistral.mixtral-8x7b-instruct-v0:1 | ❌ |
124
+ | Mistral-Large | mistral.mistral-large-2402-v1:0 | ❌ |
125
+
121
126
  To return the list progrmatically you can import and call `listBedrockWrapperSupportedModels`:
122
127
  ```javascript
123
128
  import { listBedrockWrapperSupportedModels } from 'bedrock-wrapper';
@@ -129,6 +134,54 @@ Please modify the `bedrock_models.js` file and submit a PR 🏆 or create an Iss
129
134
 
130
135
  ---
131
136
 
137
+ ### Image Support
138
+
139
+ For models with image support (Claude 3.5 Sonnet, Claude 3.7 Sonnet, and Claude 3.7 Sonnet Thinking), you can include images in your messages using the following format:
140
+
141
+ ```javascript
142
+ messages = [
143
+ {
144
+ role: "system",
145
+ content: "You are a helpful AI assistant that can analyze images.",
146
+ },
147
+ {
148
+ role: "user",
149
+ content: [
150
+ { type: "text", text: "What's in this image?" },
151
+ {
152
+ type: "image_url",
153
+ image_url: {
154
+ url: "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQEA..." // base64 encoded image
155
+ }
156
+ }
157
+ ]
158
+ }
159
+ ]
160
+ ```
161
+
162
+ You can also use a direct URL to an image instead of base64 encoding:
163
+
164
+ ```javascript
165
+ messages = [
166
+ {
167
+ role: "user",
168
+ content: [
169
+ { type: "text", text: "Describe this image in detail." },
170
+ {
171
+ type: "image_url",
172
+ image_url: {
173
+ url: "https://example.com/path/to/image.jpg" // direct URL to image
174
+ }
175
+ }
176
+ ]
177
+ }
178
+ ]
179
+ ```
180
+
181
+ You can include multiple images in a single message by adding more image_url objects to the content array.
182
+
183
+ ---
184
+
132
185
  ### 📢 P.S.
133
186
 
134
187
  In case you missed it at the beginning of this doc, for an even easier setup, use the 🔀 [Bedrock Proxy Endpoint](https://github.com/jparkerweb/bedrock-proxy-endpoint) project to spin up your own custom OpenAI server endpoint (using the standard `baseUrl`, and `apiKey` params).