@ai-sdk/provider 1.1.3 → 2.0.0-alpha.2

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.
package/CHANGELOG.md CHANGED
@@ -1,22 +1,173 @@
1
1
  # @ai-sdk/provider
2
2
 
3
- ## 1.1.3
3
+ ## 2.0.0-alpha.2
4
4
 
5
5
  ### Patch Changes
6
6
 
7
- - beef951: feat: add speech with experimental_generateSpeech
7
+ - 26535e0: release alpha.2
8
8
 
9
- ## 1.1.2
9
+ ## 2.0.0-alpha.1
10
10
 
11
11
  ### Patch Changes
12
12
 
13
- - 013faa8: core (ai): change transcription model mimeType to mediaType
13
+ - 3f2f00c: feat: `ImageModelV2#maxImagesPerCall` can be set to a function that returns a `number` or `undefined`, optionally as a promise
14
14
 
15
- ## 1.1.1
15
+ pull request: https://github.com/vercel/ai/pull/6343
16
+
17
+ ## 2.0.0-canary.14
18
+
19
+ ### Major Changes
20
+
21
+ - 7979f7f: feat (provider): support reasoning tokens, cached input tokens, total token in usage information
22
+
23
+ ### Patch Changes
24
+
25
+ - a8c8bd5: feat(embed-many): respect supportsParallelCalls & concurrency
26
+
27
+ ## 2.0.0-canary.13
28
+
29
+ ### Patch Changes
30
+
31
+ - 9bd5ab5: feat (provider): add providerMetadata to ImageModelV2 interface (#5977)
32
+
33
+ The `experimental_generateImage` method from the `ai` package now returnes revised prompts for OpenAI's image models.
34
+
35
+ ```js
36
+ const prompt = 'Santa Claus driving a Cadillac';
37
+
38
+ const { providerMetadata } = await experimental_generateImage({
39
+ model: openai.image('dall-e-3'),
40
+ prompt,
41
+ });
42
+
43
+ const revisedPrompt = providerMetadata.openai.images[0]?.revisedPrompt;
44
+
45
+ console.log({
46
+ prompt,
47
+ revisedPrompt,
48
+ });
49
+ ```
50
+
51
+ ## 2.0.0-canary.12
52
+
53
+ ### Patch Changes
54
+
55
+ - 7b3ae3f: chore (provider): change getSupportedUrls to supportedUrls (language model v2)
56
+
57
+ ## 2.0.0-canary.11
58
+
59
+ ### Major Changes
60
+
61
+ - e030615: chore (provider): remove prompt type from language model v2 spec
16
62
 
17
63
  ### Patch Changes
18
64
 
19
- - c21fa6d: feat: add transcription with experimental_transcribe
65
+ - 9bf7291: chore(providers/openai): enable structuredOutputs by default & switch to provider option
66
+ - 4617fab: chore(embedding-models): remove remaining settings
67
+
68
+ ## 2.0.0-canary.10
69
+
70
+ ### Major Changes
71
+
72
+ - a3f768e: chore: restructure reasoning support
73
+
74
+ ### Patch Changes
75
+
76
+ - 9301f86: refactor (image-model): rename `ImageModelV1` to `ImageModelV2`
77
+
78
+ ## 2.0.0-canary.9
79
+
80
+ ### Major Changes
81
+
82
+ - e86be6f: chore: remove logprobs
83
+
84
+ ## 2.0.0-canary.8
85
+
86
+ ### Major Changes
87
+
88
+ - 95857aa: chore: restructure language model supported urls
89
+ - 7ea4132: chore: remove object generation mode
90
+
91
+ ## 2.0.0-canary.7
92
+
93
+ ### Major Changes
94
+
95
+ - b6b43c7: chore: move warnings into stream-start part (spec)
96
+ - 3795467: chore: return content array from doGenerate (spec)
97
+
98
+ ### Patch Changes
99
+
100
+ - 8aa9e20: feat: add speech with experimental_generateSpeech
101
+
102
+ ## 2.0.0-canary.6
103
+
104
+ ### Major Changes
105
+
106
+ - 14c9410: chore: refactor file towards source pattern (spec)
107
+ - d9c98f4: chore: refactor reasoning parts (spec)
108
+ - 0054544: chore: refactor source parts (spec)
109
+ - 9e9c809: chore: refactor tool call and tool call delta parts (spec)
110
+ - 32831c6: chore: refactor text parts (spec)
111
+ - d0f9495: chore: refactor file parts (spec)
112
+
113
+ ### Patch Changes
114
+
115
+ - 26735b5: chore(embedding-model): add v2 interface
116
+ - 443d8ec: feat(embedding-model-v2): add response body field
117
+ - c4a2fec: chore (provider): extract shared provider options and metadata (spec)
118
+ - fd65bc6: chore(embedding-model-v2): rename rawResponse to response
119
+ - 393138b: feat(embedding-model-v2): add providerOptions
120
+ - 7182d14: Remove `Experimental_LanguageModelV2Middleware` type
121
+
122
+ ## 2.0.0-canary.5
123
+
124
+ ### Major Changes
125
+
126
+ - 411e483: chore (provider): refactor usage (language model v2)
127
+ - ad80501: chore (provider): allow both binary and base64 file content (spec)
128
+ - 1766ede: chore: rename maxTokens to maxOutputTokens
129
+
130
+ ### Patch Changes
131
+
132
+ - 79457bd: chore (provider): extract LanguageModelV2File
133
+ - f10304b: feat(tool-calling): don't require the user to have to pass parameters
134
+
135
+ ## 2.0.0-canary.4
136
+
137
+ ### Major Changes
138
+
139
+ - 6f6bb89: chore (provider): cleanup request and rawRequest (language model v2)
140
+
141
+ ## 2.0.0-canary.3
142
+
143
+ ### Major Changes
144
+
145
+ - d1a1aa1: chore (provider): merge rawRequest into request (language model v2)
146
+
147
+ ## 2.0.0-canary.2
148
+
149
+ ### Major Changes
150
+
151
+ - abf9a79: chore: rename mimeType to mediaType
152
+ - 6dc848c: chore (provider): remove image parts
153
+
154
+ ### Patch Changes
155
+
156
+ - a166433: feat: add transcription with experimental_transcribe
157
+ - 0a87932: core (ai): change transcription model mimeType to mediaType
158
+
159
+ ## 2.0.0-canary.1
160
+
161
+ ### Major Changes
162
+
163
+ - c57e248: chore (provider): remove mode
164
+ - 33f4a6a: chore (provider): rename providerMetadata inputs to providerOptions
165
+
166
+ ## 2.0.0-canary.0
167
+
168
+ ### Major Changes
169
+
170
+ - d5f588f: AI SDK 5
20
171
 
21
172
  ## 1.1.0
22
173