bedrock-wrapper 2.7.2 → 2.8.0

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 (8) hide show
  1. package/CHANGELOG.md +84 -4
  2. package/README.md +52 -32
  3. package/bedrock-models.js +353 -11
  4. package/bedrock-wrapper.js +22 -6
  5. package/package.json +2 -2
  6. package/test-converse-output.txt +7477 -0
  7. package/test-final-output.txt +7577 -0
  8. package/test-run-output.txt +7629 -0
package/CHANGELOG.md CHANGED
@@ -1,8 +1,46 @@
1
1
  # Changelog
2
+
2
3
  All notable changes to this project will be documented in this file.
3
4
 
5
+ ## [2.8.0] - 2025-12-05 (New Models: Claude Opus 4.5, Gemma, Kimi, MiniMax, Mistral, Nova)
6
+
7
+ ### ✨ Added
8
+
9
+ - Support for Claude Opus 4.5 models
10
+ - Claude-4-5-Opus (128K max output tokens, vision support)
11
+ - Claude-4-5-Opus-Thinking (with extended thinking capabilities)
12
+ - Support for Amazon Nova 2 Lite model
13
+ - Nova-2-Lite (vision support, 5K max output tokens)
14
+ - Support for Qwen3 Next model
15
+ - Qwen3-Next-80B-A3B (MoE architecture, 32K max output tokens)
16
+ - Support for new Mistral models (Converse API)
17
+ - Mistral-Large-3 (675B parameters, vision support, 32K max output tokens)
18
+ - Ministral-3-3b (vision support, 8K max output tokens)
19
+ - Ministral-3-8b (vision support, 8K max output tokens)
20
+ - Ministral-3-14b (vision support, 16K max output tokens)
21
+ - Magistral-Small-2509 (text-only, 8K max output tokens)
22
+ - Support for Google Gemma 3 models (new provider)
23
+ - Gemma-3-4b (vision support, 8K max output tokens)
24
+ - Gemma-3-12b (vision support, 8K max output tokens)
25
+ - Gemma-3-27b (vision support, 8K max output tokens)
26
+ - Support for Moonshot AI Kimi K2 models (new provider)
27
+ - Kimi-K2 (1T total parameters, 32B active MoE, 32K max output tokens)
28
+ - Kimi-K2-Thinking (with reasoning tag preservation)
29
+ - Support for MiniMax M2 model (new provider)
30
+ - MiniMax-M2 (230B total parameters, 10B active MoE, 32K max output tokens)
31
+
32
+ ### ⚙️ Technical Details
33
+
34
+ - **New Model Families**: Google Gemma, Moonshot AI Kimi, MiniMax
35
+ - **Vision Support**: All Gemma 3 models, Mistral-Large-3, Ministral 3 series, Nova-2-Lite
36
+ - **Thinking Mode**: Kimi-K2-Thinking uses `preserve_reasoning: true` for reasoning tag preservation
37
+ - **API Compatibility**: All new models use Converse API (`messages_api: true`)
38
+ - **New Mistral Models**: Unlike older Mistral models (Invoke API), new models use Converse API
39
+
4
40
  ## [2.7.0] - 2025-11-18 (DeepSeek & Qwen 3)
41
+
5
42
  ### ✨ Added
43
+
6
44
  - Support for DeepSeek foundation models
7
45
  - DeepSeek-R1 (reasoning model with chain-of-thought capabilities, 8K max output tokens)
8
46
  - DeepSeek-V3.1 (hybrid thinking mode for complex reasoning, 8K max output tokens, **Converse API only**)
@@ -21,10 +59,12 @@ All notable changes to this project will be documented in this file.
21
59
  - Repository-scale code analysis capabilities for Qwen Coder models
22
60
 
23
61
  ### 🤬 Breaking Changes
62
+
24
63
  - Removed `top_p` parameter from all models as it is not fully supported by AWS Bedrock
25
64
  - `temperature` should always be used instead
26
65
 
27
66
  ### ⚙️ Technical Details
67
+
28
68
  - **Model Configuration**: All new models use messages API format (OpenAI-compatible)
29
69
  - **API Compatibility**:
30
70
  - Qwen 3 models: Support both Invoke API and Converse API
@@ -32,7 +72,9 @@ All notable changes to this project will be documented in this file.
32
72
  - DeepSeek-V3.1: Converse API only (automatically enforced)
33
73
 
34
74
  ## [2.6.2] - 2025-10-16 (Claude Haiku 4.5)
75
+
35
76
  ### ✨ Added
77
+
36
78
  - Support for Claude Haiku 4.5 models
37
79
  - Claude-4-5-Haiku
38
80
  - Claude-4-5-Haiku-Thinking
@@ -42,16 +84,21 @@ All notable changes to this project will be documented in this file.
42
84
  - Temperature/Top-P mutual exclusion parameter handling for Haiku 4.5 models
43
85
 
44
86
  ## [2.6.1] - 2025-09-30 (Claude Sonnet 4.5)
87
+
45
88
  ### ✨ Added
89
+
46
90
  - Support for Claude Sonnet 4.5 models
47
91
  - Claude-4-5-Sonnet
48
92
  - Claude-4-5-Sonnet-Thinking
49
93
 
50
94
  ## [2.5.0] - 2025-08-12 (Converse API)
95
+
51
96
  ### ✨ Added
97
+
52
98
  - Support for Converse API (streaming and non-streaming)
53
99
 
54
100
  ### ⚙️ Technical Details
101
+
55
102
  - **Model Configuration**: All models use standard messages API format
56
103
  - **API Compatibility**: Supports OpenAI-style requests
57
104
  - **Response Processing**: Automatic reasoning tag handling based on model variant
@@ -59,7 +106,9 @@ All notable changes to this project will be documented in this file.
59
106
  - **Testing Coverage**: Full integration with existing test suites and interactive example
60
107
 
61
108
  ## [2.4.5] - 2025-08-06 (GPT-OSS Models)
109
+
62
110
  ### ✨ Added
111
+
63
112
  - Support for OpenAI GPT-OSS models on AWS Bedrock
64
113
  - GPT-OSS-120B (120B parameter open weight model)
65
114
  - GPT-OSS-20B (20B parameter open weight model)
@@ -72,6 +121,7 @@ All notable changes to this project will be documented in this file.
72
121
  - OpenAI-compatible API format with `max_completion_tokens` parameter
73
122
 
74
123
  ### ⚙️ Technical Details
124
+
75
125
  - **Model Configuration**: All GPT-OSS models use standard messages API format
76
126
  - **API Compatibility**: Supports OpenAI-style requests with Apache 2.0 licensed models
77
127
  - **Response Processing**: Automatic reasoning tag handling based on model variant
@@ -79,13 +129,17 @@ All notable changes to this project will be documented in this file.
79
129
  - **Testing Coverage**: Full integration with existing test suites and interactive example
80
130
 
81
131
  ## [2.4.4] - 2025-08-05 (Claude 4.1 Opus)
132
+
82
133
  ### ✨ Added
134
+
83
135
  - Support for Claude 4.1 Opus models
84
136
  - Claude-4-1-Opus
85
137
  - Claude-4-1-Opus-Thinking
86
138
 
87
139
  ## [2.4.3] - 2025-07-31 (Stop Sequences Fixes)
140
+
88
141
  ### 🛠️ Fixed
142
+
89
143
  - **Critical Discovery**: Removed stop sequences support from Llama models
90
144
  - AWS Bedrock does not support stop sequences for Llama models (confirmed via official AWS documentation)
91
145
  - Llama models only support: `prompt`, `temperature`, `top_p`, `max_gen_len`, `images`
@@ -95,24 +149,28 @@ All notable changes to this project will be documented in this file.
95
149
  - Improved error handling for empty responses when stop sequences trigger early
96
150
 
97
151
  ### 📝 Updated
152
+
98
153
  - **Documentation corrections**
99
154
  - Corrected stop sequences support claims (removed "all models support" language)
100
155
  - Added accurate model-specific support matrix with sequence limits
101
156
  - Added comprehensive stop sequences support table with AWS documentation references
102
157
  - **Model Support Matrix** now clearly documented:
103
- - ✅ Claude models: Full support (up to 8,191 sequences)
158
+ - ✅ Claude models: Full support (up to 8,191 sequences)
104
159
  - ✅ Nova models: Full support (up to 4 sequences)
105
160
  - ✅ Mistral models: Full support (up to 10 sequences)
106
161
  - ❌ Llama models: Not supported (AWS Bedrock limitation)
107
162
 
108
163
  ### ⚙️ Technical Details
164
+
109
165
  - Based on comprehensive research of official AWS Bedrock documentation
110
166
  - All changes maintain full backward compatibility
111
167
  - Test results show significant improvements in stop sequences reliability for supported models
112
168
  - Added detailed explanations to help users understand AWS Bedrock's actual capabilities
113
169
 
114
170
  ## [2.4.2] - 2025-07-31 (Stop Sequences Support)
171
+
115
172
  ### ✨ Added
173
+
116
174
  - Stop sequences support for compatible models
117
175
  - OpenAI-compatible `stop` and `stop_sequences` parameters
118
176
  - Automatic string-to-array conversion for compatibility
@@ -121,6 +179,7 @@ All notable changes to this project will be documented in this file.
121
179
  - Comprehensive stop sequences testing and validation with `npm run test-stop`
122
180
 
123
181
  ### 🛠️ Fixed
182
+
124
183
  - **Critical Discovery**: Removed stop sequences support from Llama models
125
184
  - AWS Bedrock does not support stop sequences for Llama models (confirmed via official documentation)
126
185
  - Llama models only support: `prompt`, `temperature`, `top_p`, `max_gen_len`, `images`
@@ -129,6 +188,7 @@ All notable changes to this project will be documented in this file.
129
188
  - Improved error handling for empty responses when stop sequences trigger early
130
189
 
131
190
  ### ⚙️ Technical Details
191
+
132
192
  - **Model Support Matrix**:
133
193
  - ✅ Claude models: Full support (up to 8,191 sequences)
134
194
  - ✅ Nova models: Full support (up to 4 sequences)
@@ -140,7 +200,9 @@ All notable changes to this project will be documented in this file.
140
200
  - Added comprehensive documentation in README.md and CLAUDE.md explaining support limitations
141
201
 
142
202
  ## [2.4.0] - 2025-07-24 (AWS Nova Models)
203
+
143
204
  ### ✨ Added
205
+
144
206
  - Support for AWS Nova models
145
207
  - Nova-Pro (300K context, multimodal, 5K output tokens)
146
208
  - Nova-Lite (300K context, multimodal, optimized for speed)
@@ -150,7 +212,9 @@ All notable changes to this project will be documented in this file.
150
212
  - Automatic content array formatting for Nova message compatibility
151
213
 
152
214
  ## [2.3.1] - 2025-05-22 (Claude 4 Opus / Sonnet)
215
+
153
216
  ### ✨ Added
217
+
154
218
  - Support for Claude 4 Opus & Claude 4 Sonnet models
155
219
  - Claude-4-Opus
156
220
  - Claude-4-Opus-Thinking
@@ -158,7 +222,9 @@ All notable changes to this project will be documented in this file.
158
222
  - Claude-4-Sonnet-Thinking
159
223
 
160
224
  ## [2.3.0] - 2025-02-15 (Claude 3.7 & Image Support)
225
+
161
226
  ### ✨ Added
227
+
162
228
  - Support for Claude 3.7 models
163
229
  - Claude-3-7-Sonnet
164
230
  - Claude-3-7-Sonnet-Thinking
@@ -171,29 +237,37 @@ All notable changes to this project will be documented in this file.
171
237
  - Documentation for image support usage
172
238
 
173
239
  ### 🔄 Changed
240
+
174
241
  - Updated model configuration for image-capable models
175
242
  - Improved response handling for multimodal inputs
176
243
 
177
244
  ## [2.2.0] - 2025-01-01 (Llama 3.3 70b)
245
+
178
246
  ### ✨ Added
247
+
179
248
  - Support for Llama 3.3 70b
180
249
 
181
250
  ## [2.1.0] - 2024-11-21 (Claude 3.5 Haiku)
251
+
182
252
  ### ✨ Added
253
+
183
254
  - Support for Claude 3.5 Haiku
184
255
 
185
256
  ## [2.0.0] - 2024-10-31 (Claude Sonnet & Haiku)
257
+
186
258
  ### ✨ Added
259
+
187
260
  - Support for Anthropic Sonnet & Haiku models
188
261
  - Claude-3-5-Sonnet-v2
189
262
  - Claude-3-5-Sonnet
190
263
  - Claude-3-Haiku
191
264
  - Interactive example script for testing models
192
265
  - Testing script with streaming and non-streaming support for all models
193
- - Stardardize output to be a string via Streamed and non-Streamed responses
266
+ - Stardardize output to be a string via Streamed and non-Streamed responses
194
267
  > **NOTE:** This is a breaking change for previous non-streaming responses. Existing streaming responses will remain unchanged.
195
268
 
196
269
  ### 🔄 Changed
270
+
197
271
  - Complete architecture overhaul for better model support
198
272
  - Improved message handling with role-based formatting
199
273
  - Enhanced error handling and response processing
@@ -201,6 +275,7 @@ All notable changes to this project will be documented in this file.
201
275
  - Updated AWS SDK integration
202
276
 
203
277
  ### ⚙️ Technical Details
278
+
204
279
  - Implemented messages API support for compatible models
205
280
  - Added system message handling as separate field where supported
206
281
  - Configurable token limits per model
@@ -208,7 +283,9 @@ All notable changes to this project will be documented in this file.
208
283
  - Cross-region profile support for certain models
209
284
 
210
285
  ## [1.3.0] - 2024-07-24 (Llama3.2)
286
+
211
287
  ### ✨ Added
288
+
212
289
  - Support for Llama 3.2 series models
213
290
  - Llama-3-2-1b
214
291
  - Llama-3-2-3b
@@ -216,18 +293,21 @@ All notable changes to this project will be documented in this file.
216
293
  - Llama-3-2-90b
217
294
 
218
295
  ## [1.1.0] - 2024-07-24 (Llama3.1)
296
+
219
297
  ### ✨ Added
298
+
220
299
  - Support for Llama 3.1 series models
221
300
  - Llama-3-1-8b
222
301
  - Llama-3-1-70b
223
302
 
224
-
225
303
  ## [1.0.14] - 2024-05-06 (Initial Stable Release)
304
+
226
305
  ### ✨ Added
306
+
227
307
  - Initial stablerelease of Bedrock Wrapper
228
308
  - Basic AWS Bedrock integration
229
309
  - OpenAI-compatible API object support
230
- - Basic model support
310
+ - Basic model support
231
311
  - Llama-3-8b
232
312
  - Llama-3-70b
233
313
  - Mistral-7b
package/README.md CHANGED
@@ -122,43 +122,57 @@ Bedrock Wrapper is an npm package that simplifies the integration of existing Op
122
122
 
123
123
  ### Supported Models
124
124
 
125
- | modelName | AWS Model Id | Image |
126
- |----------------------------|----------------------------------------------|-------|
127
- | Claude-3-5-Haiku | anthropic.claude-3-5-haiku-20241022-v1:0 | ❌ |
128
- | Claude-3-5-Sonnet | anthropic.claude-3-5-sonnet-20240620-v1:0 | ✅ |
129
- | Claude-3-5-Sonnet-v2 | anthropic.claude-3-5-sonnet-20241022-v2:0 | ✅ |
130
- | Claude-3-7-Sonnet | us.anthropic.claude-3-7-sonnet-20250219-v1:0 | ✅ |
131
- | Claude-3-7-Sonnet-Thinking | us.anthropic.claude-3-7-sonnet-20250219-v1:0 | ✅ |
132
- | Claude-3-Haiku | anthropic.claude-3-haiku-20240307-v1:0 | ✅ |
133
- | Claude-4-Opus | us.anthropic.claude-opus-4-20250514-v1:0 | ✅ |
134
- | Claude-4-Opus-Thinking | us.anthropic.claude-opus-4-20250514-v1:0 | ✅ |
135
- | Claude-4-Sonnet | us.anthropic.claude-sonnet-4-20250514-v1:0 | ✅ |
136
- | Claude-4-Sonnet-Thinking | us.anthropic.claude-sonnet-4-20250514-v1:0 | ✅ |
137
- | Claude-4-1-Opus | us.anthropic.claude-opus-4-1-20250805-v1:0 | ✅ |
138
- | Claude-4-1-Opus-Thinking | us.anthropic.claude-opus-4-1-20250805-v1:0 | ✅ |
139
- | Claude-4-5-Haiku | us.anthropic.claude-haiku-4-5-20251001-v1:0 | ✅ |
140
- | Claude-4-5-Haiku-Thinking | us.anthropic.claude-haiku-4-5-20251001-v1:0 | ✅ |
141
- | Claude-4-5-Sonnet | us.anthropic.claude-sonnet-4-5-20250929-v1:0 | ✅ |
142
- | Claude-4-5-Sonnet-Thinking | us.anthropic.claude-sonnet-4-5-20250929-v1:0 | ✅ |
143
- | DeepSeek-R1 | us.deepseek.r1-v1:0 | ❌ |
144
- | DeepSeek-V3.1 | deepseek.v3-v1:0 | ❌ |
145
- | GPT-OSS-120B | openai.gpt-oss-120b-1:0 | ❌ |
146
- | GPT-OSS-120B-Thinking | openai.gpt-oss-120b-1:0 | ❌ |
147
- | GPT-OSS-20B | openai.gpt-oss-20b-1:0 | ❌ |
148
- | GPT-OSS-20B-Thinking | openai.gpt-oss-20b-1:0 | ❌ |
149
- | Llama-3-8b | meta.llama3-8b-instruct-v1:0 | ❌ |
150
- | Llama-3-70b | meta.llama3-70b-instruct-v1:0 | ❌ |
151
- | Llama-3-1-8b | meta.llama3-1-8b-instruct-v1:0 | ❌ |
152
- | Llama-3-1-70b | meta.llama3-1-70b-instruct-v1:0 | ❌ |
153
- | Llama-3-1-405b | meta.llama3-1-405b-instruct-v1:0 | ❌ |
125
+ | modelName | AWS Model Id | Image |
126
+ |----------------------------|-------------------------------------------------|-------|
127
+ | Claude-3-5-Haiku | us.anthropic.claude-3-5-haiku-20241022-v1:0 | ❌ |
128
+ | Claude-3-5-Sonnet | us.anthropic.claude-3-5-sonnet-20240620-v1:0 | ✅ |
129
+ | Claude-3-5-Sonnet-v2 | us.anthropic.claude-3-5-sonnet-20241022-v2:0 | ✅ |
130
+ | Claude-3-7-Sonnet | us.anthropic.claude-3-7-sonnet-20250219-v1:0 | ✅ |
131
+ | Claude-3-7-Sonnet-Thinking | us.anthropic.claude-3-7-sonnet-20250219-v1:0 | ✅ |
132
+ | Claude-3-Haiku | us.anthropic.claude-3-haiku-20240307-v1:0 | ✅ |
133
+ | Claude-4-Opus | us.anthropic.claude-opus-4-20250514-v1:0 | ✅ |
134
+ | Claude-4-Opus-Thinking | us.anthropic.claude-opus-4-20250514-v1:0 | ✅ |
135
+ | Claude-4-Sonnet | us.anthropic.claude-sonnet-4-20250514-v1:0 | ✅ |
136
+ | Claude-4-Sonnet-Thinking | us.anthropic.claude-sonnet-4-20250514-v1:0 | ✅ |
137
+ | Claude-4-1-Opus | us.anthropic.claude-opus-4-1-20250805-v1:0 | ✅ |
138
+ | Claude-4-1-Opus-Thinking | us.anthropic.claude-opus-4-1-20250805-v1:0 | ✅ |
139
+ | Claude-4-5-Haiku | global.anthropic.claude-haiku-4-5-20251001-v1:0 | ✅ |
140
+ | Claude-4-5-Haiku-Thinking | global.anthropic.claude-haiku-4-5-20251001-v1:0 | ✅ |
141
+ | Claude-4-5-Opus | global.anthropic.claude-opus-4-5-20251101-v1:0 | ✅ |
142
+ | Claude-4-5-Opus-Thinking | global.anthropic.claude-opus-4-5-20251101-v1:0 | ✅ |
143
+ | Claude-4-5-Sonnet | us.anthropic.claude-sonnet-4-5-20250929-v1:0 | ✅ |
144
+ | Claude-4-5-Sonnet-Thinking | us.anthropic.claude-sonnet-4-5-20250929-v1:0 | ✅ |
145
+ | DeepSeek-R1 | us.deepseek.r1-v1:0 | ❌ |
146
+ | DeepSeek-V3.1 | deepseek.v3-v1:0 | ❌ |
147
+ | Gemma-3-4b | google.gemma-3-4b-it | ✅ |
148
+ | Gemma-3-12b | google.gemma-3-12b-it | ✅ |
149
+ | Gemma-3-27b | google.gemma-3-27b-it | ✅ |
150
+ | GPT-OSS-120B | openai.gpt-oss-120b-1:0 | ❌ |
151
+ | GPT-OSS-120B-Thinking | openai.gpt-oss-120b-1:0 | ❌ |
152
+ | GPT-OSS-20B | openai.gpt-oss-20b-1:0 | ❌ |
153
+ | GPT-OSS-20B-Thinking | openai.gpt-oss-20b-1:0 | ❌ |
154
+ | Kimi-K2 | moonshot.kimi-k2-thinking | ❌ |
155
+ | Kimi-K2-Thinking | moonshot.kimi-k2-thinking | ❌ |
156
+ | Llama-3-8b | meta.llama3-8b-instruct-v1:0 | ❌ |
157
+ | Llama-3-70b | meta.llama3-70b-instruct-v1:0 | ❌ |
158
+ | Llama-3-1-8b | us.meta.llama3-1-8b-instruct-v1:0 | ❌ |
159
+ | Llama-3-1-70b | us.meta.llama3-1-70b-instruct-v1:0 | ❌ |
160
+ | Llama-3-1-405b | meta.llama3-1-405b-instruct-v1:0 | ❌ |
154
161
  | Llama-3-2-1b | us.meta.llama3-2-1b-instruct-v1:0 | ❌ |
155
162
  | Llama-3-2-3b | us.meta.llama3-2-3b-instruct-v1:0 | ❌ |
156
163
  | Llama-3-2-11b | us.meta.llama3-2-11b-instruct-v1:0 | ❌ |
157
164
  | Llama-3-2-90b | us.meta.llama3-2-90b-instruct-v1:0 | ❌ |
158
165
  | Llama-3-3-70b | us.meta.llama3-3-70b-instruct-v1:0 | ❌ |
166
+ | Magistral-Small-2509 | mistral.magistral-small-2509 | ❌ |
167
+ | MiniMax-M2 | minimax.minimax-m2 | ❌ |
168
+ | Ministral-3-3b | mistral.ministral-3-3b-instruct | ✅ |
169
+ | Ministral-3-8b | mistral.ministral-3-8b-instruct | ✅ |
170
+ | Ministral-3-14b | mistral.ministral-3-14b-instruct | ✅ |
159
171
  | Mistral-7b | mistral.mistral-7b-instruct-v0:2 | ❌ |
160
- | Mixtral-8x7b | mistral.mixtral-8x7b-instruct-v0:1 | ❌ |
161
172
  | Mistral-Large | mistral.mistral-large-2402-v1:0 | ❌ |
173
+ | Mistral-Large-3 | mistral.mistral-large-3-675b-instruct | ✅ |
174
+ | Mixtral-8x7b | mistral.mixtral-8x7b-instruct-v0:1 | ❌ |
175
+ | Nova-2-Lite | us.amazon.nova-2-lite-v1:0 | ✅ |
162
176
  | Nova-Micro | us.amazon.nova-micro-v1:0 | ❌ |
163
177
  | Nova-Lite | us.amazon.nova-lite-v1:0 | ✅ |
164
178
  | Nova-Pro | us.amazon.nova-pro-v1:0 | ✅ |
@@ -166,6 +180,7 @@ Bedrock Wrapper is an npm package that simplifies the integration of existing Op
166
180
  | Qwen3-235B-A22B-2507 | qwen.qwen3-235b-a22b-2507-v1:0 | ❌ |
167
181
  | Qwen3-Coder-30B-A3B | qwen.qwen3-coder-30b-a3b-v1:0 | ❌ |
168
182
  | Qwen3-Coder-480B-A35B | qwen.qwen3-coder-480b-a35b-v1:0 | ❌ |
183
+ | Qwen3-Next-80B-A3B | qwen.qwen3-next-80b-a3b | ❌ |
169
184
 
170
185
  To return the list progrmatically you can import and call `listBedrockWrapperSupportedModels`:
171
186
  ```javascript
@@ -181,8 +196,9 @@ Please modify the `bedrock_models.js` file and submit a PR 🏆 or create an Iss
181
196
  ### Thinking Models
182
197
 
183
198
  Some models support extended reasoning capabilities through "thinking mode". These models include:
184
- - **Claude models**: Claude-4-1-Opus-Thinking, Claude-4-Opus-Thinking, Claude-4-5-Sonnet-Thinking, Claude-4-5-Haiku-Thinking, Claude-4-Sonnet-Thinking, Claude-3-7-Sonnet-Thinking
199
+ - **Claude models**: Claude-4-5-Opus-Thinking, Claude-4-1-Opus-Thinking, Claude-4-Opus-Thinking, Claude-4-5-Sonnet-Thinking, Claude-4-5-Haiku-Thinking, Claude-4-Sonnet-Thinking, Claude-3-7-Sonnet-Thinking
185
200
  - **GPT-OSS models**: GPT-OSS-120B-Thinking, GPT-OSS-20B-Thinking
201
+ - **Kimi models**: Kimi-K2-Thinking (preserves reasoning tags in output)
186
202
 
187
203
  To use thinking mode and see the model's reasoning process, set `include_thinking_data: true` in your request:
188
204
 
@@ -213,7 +229,7 @@ for await (const chunk of bedrockWrapper(awsCreds, openaiChatCompletionsCreateOb
213
229
 
214
230
  ### Image Support
215
231
 
216
- For models with image support (Claude 4+ series including Claude 4.5 Sonnet, Claude 4.5 Haiku, Claude 3.7 Sonnet, Claude 3.5 Sonnet, Claude 3 Haiku, Nova Pro, and Nova Lite), you can include images in your messages using the following format (not all models support system prompts):
232
+ For models with image support (Claude 4+ series including Claude 4.5 Opus, Claude 4.5 Sonnet, Claude 4.5 Haiku, Claude 3.7 Sonnet, Claude 3.5 Sonnet, Claude 3 Haiku, Nova Pro, Nova Lite, Nova 2 Lite, Mistral Large 3, Ministral 3 series, and Gemma 3 series), you can include images in your messages using the following format (not all models support system prompts):
217
233
 
218
234
  ```javascript
219
235
  messages = [
@@ -280,6 +296,9 @@ const openaiChatCompletionsCreateObject = {
280
296
  - ✅ **GPT-OSS models**: Fully supported
281
297
  - ✅ **Mistral models**: Fully supported (up to 10 sequences)
282
298
  - ✅ **Qwen models**: Fully supported
299
+ - ✅ **Gemma models**: Fully supported
300
+ - ✅ **Kimi models**: Fully supported
301
+ - ✅ **MiniMax models**: Fully supported
283
302
  - ❌ **Llama models**: Not supported (AWS Bedrock limitation)
284
303
 
285
304
  **Features:**
@@ -310,6 +329,7 @@ Some AWS Bedrock models have specific parameter restrictions that are automatica
310
329
  #### Claude 4+ Models (Temperature/Top-P Mutual Exclusion)
311
330
 
312
331
  **Affected Models:**
332
+ - Claude-4-5-Opus & Claude-4-5-Opus-Thinking
313
333
  - Claude-4-5-Sonnet & Claude-4-5-Sonnet-Thinking
314
334
  - Claude-4-5-Haiku & Claude-4-5-Haiku-Thinking
315
335
  - Claude-4-Sonnet & Claude-4-Sonnet-Thinking