retab 0.0.35__py3-none-any.whl
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.
- retab-0.0.35.dist-info/METADATA +417 -0
- retab-0.0.35.dist-info/RECORD +111 -0
- retab-0.0.35.dist-info/WHEEL +5 -0
- retab-0.0.35.dist-info/top_level.txt +1 -0
- uiform/__init__.py +4 -0
- uiform/_resource.py +28 -0
- uiform/_utils/__init__.py +0 -0
- uiform/_utils/ai_models.py +100 -0
- uiform/_utils/benchmarking copy.py +588 -0
- uiform/_utils/benchmarking.py +485 -0
- uiform/_utils/chat.py +332 -0
- uiform/_utils/display.py +443 -0
- uiform/_utils/json_schema.py +2161 -0
- uiform/_utils/mime.py +168 -0
- uiform/_utils/responses.py +163 -0
- uiform/_utils/stream_context_managers.py +52 -0
- uiform/_utils/usage/__init__.py +0 -0
- uiform/_utils/usage/usage.py +300 -0
- uiform/client.py +701 -0
- uiform/py.typed +0 -0
- uiform/resources/__init__.py +0 -0
- uiform/resources/consensus/__init__.py +3 -0
- uiform/resources/consensus/client.py +114 -0
- uiform/resources/consensus/completions.py +252 -0
- uiform/resources/consensus/completions_stream.py +278 -0
- uiform/resources/consensus/responses.py +325 -0
- uiform/resources/consensus/responses_stream.py +373 -0
- uiform/resources/deployments/__init__.py +9 -0
- uiform/resources/deployments/client.py +78 -0
- uiform/resources/deployments/endpoints.py +322 -0
- uiform/resources/deployments/links.py +452 -0
- uiform/resources/deployments/logs.py +211 -0
- uiform/resources/deployments/mailboxes.py +496 -0
- uiform/resources/deployments/outlook.py +531 -0
- uiform/resources/deployments/tests.py +158 -0
- uiform/resources/documents/__init__.py +3 -0
- uiform/resources/documents/client.py +255 -0
- uiform/resources/documents/extractions.py +441 -0
- uiform/resources/evals.py +812 -0
- uiform/resources/files.py +24 -0
- uiform/resources/finetuning.py +62 -0
- uiform/resources/jsonlUtils.py +1046 -0
- uiform/resources/models.py +45 -0
- uiform/resources/openai_example.py +22 -0
- uiform/resources/processors/__init__.py +3 -0
- uiform/resources/processors/automations/__init__.py +9 -0
- uiform/resources/processors/automations/client.py +78 -0
- uiform/resources/processors/automations/endpoints.py +317 -0
- uiform/resources/processors/automations/links.py +356 -0
- uiform/resources/processors/automations/logs.py +211 -0
- uiform/resources/processors/automations/mailboxes.py +435 -0
- uiform/resources/processors/automations/outlook.py +444 -0
- uiform/resources/processors/automations/tests.py +158 -0
- uiform/resources/processors/client.py +474 -0
- uiform/resources/prompt_optimization.py +76 -0
- uiform/resources/schemas.py +369 -0
- uiform/resources/secrets/__init__.py +9 -0
- uiform/resources/secrets/client.py +20 -0
- uiform/resources/secrets/external_api_keys.py +109 -0
- uiform/resources/secrets/webhook.py +62 -0
- uiform/resources/usage.py +271 -0
- uiform/types/__init__.py +0 -0
- uiform/types/ai_models.py +645 -0
- uiform/types/automations/__init__.py +0 -0
- uiform/types/automations/cron.py +58 -0
- uiform/types/automations/endpoints.py +21 -0
- uiform/types/automations/links.py +28 -0
- uiform/types/automations/mailboxes.py +60 -0
- uiform/types/automations/outlook.py +68 -0
- uiform/types/automations/webhooks.py +21 -0
- uiform/types/chat.py +8 -0
- uiform/types/completions.py +93 -0
- uiform/types/consensus.py +10 -0
- uiform/types/db/__init__.py +0 -0
- uiform/types/db/annotations.py +24 -0
- uiform/types/db/files.py +36 -0
- uiform/types/deployments/__init__.py +0 -0
- uiform/types/deployments/cron.py +59 -0
- uiform/types/deployments/endpoints.py +28 -0
- uiform/types/deployments/links.py +36 -0
- uiform/types/deployments/mailboxes.py +67 -0
- uiform/types/deployments/outlook.py +76 -0
- uiform/types/deployments/webhooks.py +21 -0
- uiform/types/documents/__init__.py +0 -0
- uiform/types/documents/correct_orientation.py +13 -0
- uiform/types/documents/create_messages.py +226 -0
- uiform/types/documents/extractions.py +297 -0
- uiform/types/evals.py +207 -0
- uiform/types/events.py +76 -0
- uiform/types/extractions.py +85 -0
- uiform/types/jobs/__init__.py +0 -0
- uiform/types/jobs/base.py +150 -0
- uiform/types/jobs/batch_annotation.py +22 -0
- uiform/types/jobs/evaluation.py +133 -0
- uiform/types/jobs/finetune.py +6 -0
- uiform/types/jobs/prompt_optimization.py +41 -0
- uiform/types/jobs/webcrawl.py +6 -0
- uiform/types/logs.py +231 -0
- uiform/types/mime.py +257 -0
- uiform/types/modalities.py +68 -0
- uiform/types/pagination.py +6 -0
- uiform/types/schemas/__init__.py +0 -0
- uiform/types/schemas/enhance.py +53 -0
- uiform/types/schemas/evaluate.py +55 -0
- uiform/types/schemas/generate.py +32 -0
- uiform/types/schemas/layout.py +58 -0
- uiform/types/schemas/object.py +631 -0
- uiform/types/schemas/templates.py +107 -0
- uiform/types/secrets/__init__.py +0 -0
- uiform/types/secrets/external_api_keys.py +22 -0
- uiform/types/standards.py +39 -0
@@ -0,0 +1,645 @@
|
|
1
|
+
from typing import Literal
|
2
|
+
|
3
|
+
AIProvider = Literal["OpenAI", "Gemini", "xAI", "UiForm"] # , "Anthropic", "xAI"]
|
4
|
+
OpenAICompatibleProvider = Literal["OpenAI", "xAI"] # , "xAI"]
|
5
|
+
GeminiModel = Literal[
|
6
|
+
"gemini-2.5-pro-preview-05-06",
|
7
|
+
"gemini-2.5-flash-preview-05-20",
|
8
|
+
"gemini-2.5-flash-preview-04-17",
|
9
|
+
"gemini-2.5-pro-exp-03-25",
|
10
|
+
"gemini-2.0-flash-lite",
|
11
|
+
"gemini-2.0-flash",
|
12
|
+
]
|
13
|
+
|
14
|
+
AnthropicModel = Literal[
|
15
|
+
"claude-3-5-sonnet-latest", "claude-3-5-sonnet-20241022", "claude-3-5-haiku-20241022", "claude-3-opus-20240229", "claude-3-sonnet-20240229", "claude-3-haiku-20240307"
|
16
|
+
]
|
17
|
+
OpenAIModel = Literal[
|
18
|
+
"gpt-4o",
|
19
|
+
"gpt-4o-mini",
|
20
|
+
"chatgpt-4o-latest",
|
21
|
+
"gpt-4.1",
|
22
|
+
"gpt-4.1-mini",
|
23
|
+
"gpt-4.1-nano",
|
24
|
+
"gpt-4o-2024-11-20",
|
25
|
+
"gpt-4o-2024-08-06",
|
26
|
+
"gpt-4o-2024-05-13",
|
27
|
+
"gpt-4o-mini-2024-07-18",
|
28
|
+
"o3-mini",
|
29
|
+
"o3-mini-2025-01-31",
|
30
|
+
"o1",
|
31
|
+
"o1-2024-12-17",
|
32
|
+
"o1-preview-2024-09-12",
|
33
|
+
"o1-mini",
|
34
|
+
"o1-mini-2024-09-12",
|
35
|
+
"o3",
|
36
|
+
"o3-2025-04-16",
|
37
|
+
"o4-mini",
|
38
|
+
"o4-mini-2025-04-16",
|
39
|
+
"gpt-4.5-preview",
|
40
|
+
"gpt-4.5-preview-2025-02-27",
|
41
|
+
"gpt-4o-audio-preview-2024-12-17",
|
42
|
+
"gpt-4o-audio-preview-2024-10-01",
|
43
|
+
"gpt-4o-realtime-preview-2024-12-17",
|
44
|
+
"gpt-4o-realtime-preview-2024-10-01",
|
45
|
+
"gpt-4o-mini-audio-preview-2024-12-17",
|
46
|
+
"gpt-4o-mini-realtime-preview-2024-12-17",
|
47
|
+
]
|
48
|
+
xAI_Model = Literal["grok-3-beta", "grok-3-mini-beta"]
|
49
|
+
UiFormModel = Literal["auto", "auto-small"]
|
50
|
+
LLMModel = Literal[OpenAIModel, "human", AnthropicModel, xAI_Model, GeminiModel, UiFormModel]
|
51
|
+
|
52
|
+
import datetime
|
53
|
+
|
54
|
+
from pydantic import BaseModel, Field
|
55
|
+
|
56
|
+
from uiform.types.jobs.base import InferenceSettings
|
57
|
+
|
58
|
+
|
59
|
+
class FinetunedModel(BaseModel):
|
60
|
+
object: Literal["finetuned_model"] = "finetuned_model"
|
61
|
+
organization_id: str
|
62
|
+
model: str
|
63
|
+
schema_id: str
|
64
|
+
schema_data_id: str
|
65
|
+
finetuning_props: InferenceSettings
|
66
|
+
evaluation_id: str | None = None
|
67
|
+
created_at: datetime.datetime = Field(default_factory=lambda: datetime.datetime.now(datetime.timezone.utc))
|
68
|
+
|
69
|
+
|
70
|
+
from typing import Dict, List, Literal, Optional
|
71
|
+
|
72
|
+
from pydantic import BaseModel
|
73
|
+
|
74
|
+
|
75
|
+
# Monthly Usage
|
76
|
+
class MonthlyUsageResponseContent(BaseModel):
|
77
|
+
request_count: int
|
78
|
+
|
79
|
+
|
80
|
+
MonthlyUsageResponse = MonthlyUsageResponseContent
|
81
|
+
|
82
|
+
|
83
|
+
class Amount(BaseModel):
|
84
|
+
value: float
|
85
|
+
currency: str
|
86
|
+
|
87
|
+
|
88
|
+
class TokenPrice(BaseModel):
|
89
|
+
"""
|
90
|
+
Holds pricing information (price per 1M tokens) for one token category.
|
91
|
+
(For example, for text tokens used in the prompt.)
|
92
|
+
"""
|
93
|
+
|
94
|
+
prompt: float # Price per 1M prompt tokens.
|
95
|
+
completion: float # Price per 1M completion tokens.
|
96
|
+
cached_discount: float = 0.5 # Price discount for cached tokens.
|
97
|
+
|
98
|
+
|
99
|
+
class Pricing(BaseModel):
|
100
|
+
"""
|
101
|
+
Contains all pricing information that is useful for a given model.
|
102
|
+
(For example, the grid below shows that for gpt-4o-2024-08-06 text prompt tokens
|
103
|
+
cost $2.50 per 1M tokens while completion tokens cost $10.00 per 1M tokens.)
|
104
|
+
"""
|
105
|
+
|
106
|
+
text: TokenPrice
|
107
|
+
audio: Optional[TokenPrice] = None # May be None if the model does not support audio tokens.
|
108
|
+
ft_price_hike: float = 1.0 # Price hike for fine-tuned models.
|
109
|
+
|
110
|
+
|
111
|
+
# Fix the type definition - use proper Python syntax
|
112
|
+
ModelModality = Literal["text", "audio", "image"]
|
113
|
+
|
114
|
+
# Define supported endpoints
|
115
|
+
EndpointType = Literal[
|
116
|
+
"chat_completions",
|
117
|
+
"responses",
|
118
|
+
"assistants",
|
119
|
+
"batch",
|
120
|
+
"fine_tuning",
|
121
|
+
"embeddings",
|
122
|
+
"speech_generation",
|
123
|
+
"translation",
|
124
|
+
"completions_legacy",
|
125
|
+
"image_generation",
|
126
|
+
"transcription",
|
127
|
+
"moderation",
|
128
|
+
"realtime",
|
129
|
+
]
|
130
|
+
|
131
|
+
# Define supported features
|
132
|
+
FeatureType = Literal["streaming", "function_calling", "structured_outputs", "distillation", "fine_tuning", "predicted_outputs"]
|
133
|
+
|
134
|
+
|
135
|
+
class ModelCapabilities(BaseModel):
|
136
|
+
"""
|
137
|
+
Represents the capabilities of an AI model, including supported modalities,
|
138
|
+
endpoints, and features.
|
139
|
+
"""
|
140
|
+
|
141
|
+
modalities: List[ModelModality]
|
142
|
+
endpoints: List[EndpointType]
|
143
|
+
features: List[FeatureType]
|
144
|
+
|
145
|
+
|
146
|
+
class ModelCard(BaseModel):
|
147
|
+
"""
|
148
|
+
Model card that includes pricing and capabilities.
|
149
|
+
"""
|
150
|
+
|
151
|
+
model: LLMModel | str # Can be a random string for finetuned models
|
152
|
+
pricing: Pricing
|
153
|
+
capabilities: ModelCapabilities
|
154
|
+
temperature_support: bool = True
|
155
|
+
reasoning_effort_support: bool = False
|
156
|
+
|
157
|
+
|
158
|
+
# List of model cards with pricing and capabilities
|
159
|
+
openai_model_cards = [
|
160
|
+
########################
|
161
|
+
########################
|
162
|
+
# ----------------------
|
163
|
+
# Reasoning models
|
164
|
+
# ----------------------
|
165
|
+
########################
|
166
|
+
########################
|
167
|
+
# ----------------------
|
168
|
+
# o1 family
|
169
|
+
# ----------------------
|
170
|
+
ModelCard(
|
171
|
+
model="o1",
|
172
|
+
pricing=Pricing(text=TokenPrice(prompt=15.00, cached_discount=0.5, completion=60.00), audio=None),
|
173
|
+
capabilities=ModelCapabilities(
|
174
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
175
|
+
),
|
176
|
+
temperature_support=False,
|
177
|
+
reasoning_effort_support=True,
|
178
|
+
),
|
179
|
+
ModelCard(
|
180
|
+
model="o1-2024-12-17",
|
181
|
+
pricing=Pricing(text=TokenPrice(prompt=15.00, cached_discount=0.5, completion=60.00), audio=None),
|
182
|
+
capabilities=ModelCapabilities(
|
183
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
184
|
+
),
|
185
|
+
temperature_support=False,
|
186
|
+
reasoning_effort_support=True,
|
187
|
+
),
|
188
|
+
ModelCard(
|
189
|
+
model="o1-preview-2024-09-12",
|
190
|
+
pricing=Pricing(text=TokenPrice(prompt=15.00, cached_discount=0.5, completion=60.00), audio=None),
|
191
|
+
capabilities=ModelCapabilities(modalities=["text"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling"]),
|
192
|
+
temperature_support=False,
|
193
|
+
reasoning_effort_support=True,
|
194
|
+
),
|
195
|
+
# ----------------------
|
196
|
+
# o1-mini family
|
197
|
+
# ----------------------
|
198
|
+
ModelCard(
|
199
|
+
model="o1-mini",
|
200
|
+
pricing=Pricing(text=TokenPrice(prompt=1.10, cached_discount=0.5, completion=4.40), audio=None),
|
201
|
+
capabilities=ModelCapabilities(modalities=["text"], endpoints=["chat_completions", "responses", "assistants"], features=["streaming"]),
|
202
|
+
temperature_support=False,
|
203
|
+
reasoning_effort_support=True,
|
204
|
+
),
|
205
|
+
ModelCard(
|
206
|
+
model="o1-mini-2024-09-12",
|
207
|
+
pricing=Pricing(text=TokenPrice(prompt=1.10, cached_discount=0.5, completion=4.40), audio=None),
|
208
|
+
capabilities=ModelCapabilities(modalities=["text"], endpoints=["chat_completions", "responses", "assistants"], features=["streaming"]),
|
209
|
+
temperature_support=False,
|
210
|
+
reasoning_effort_support=True,
|
211
|
+
),
|
212
|
+
# ----------------------
|
213
|
+
# o3 family
|
214
|
+
# ----------------------
|
215
|
+
ModelCard(
|
216
|
+
model="o3",
|
217
|
+
pricing=Pricing(text=TokenPrice(prompt=10.0, cached_discount=2.5 / 10.0, completion=40.0), audio=None, ft_price_hike=1.5),
|
218
|
+
capabilities=ModelCapabilities(
|
219
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
220
|
+
),
|
221
|
+
temperature_support=False,
|
222
|
+
reasoning_effort_support=True,
|
223
|
+
),
|
224
|
+
ModelCard(
|
225
|
+
model="o3-2025-04-16",
|
226
|
+
pricing=Pricing(text=TokenPrice(prompt=10.0, cached_discount=2.5 / 10.0, completion=40.0), audio=None, ft_price_hike=1.5),
|
227
|
+
capabilities=ModelCapabilities(
|
228
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
229
|
+
),
|
230
|
+
temperature_support=False,
|
231
|
+
reasoning_effort_support=True,
|
232
|
+
),
|
233
|
+
# ----------------------
|
234
|
+
# o3-mini family
|
235
|
+
# ----------------------
|
236
|
+
ModelCard(
|
237
|
+
model="o3-mini-2025-01-31",
|
238
|
+
pricing=Pricing(text=TokenPrice(prompt=1.10, cached_discount=0.5, completion=4.40), audio=None, ft_price_hike=1.5),
|
239
|
+
capabilities=ModelCapabilities(
|
240
|
+
modalities=["text"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
241
|
+
),
|
242
|
+
temperature_support=False,
|
243
|
+
reasoning_effort_support=True,
|
244
|
+
),
|
245
|
+
ModelCard(
|
246
|
+
model="o3-mini",
|
247
|
+
pricing=Pricing(text=TokenPrice(prompt=1.10, cached_discount=0.5, completion=4.40), audio=None, ft_price_hike=1.5),
|
248
|
+
capabilities=ModelCapabilities(
|
249
|
+
modalities=["text"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
250
|
+
),
|
251
|
+
temperature_support=False,
|
252
|
+
reasoning_effort_support=True,
|
253
|
+
),
|
254
|
+
# ----------------------
|
255
|
+
# o4-mini family
|
256
|
+
# ----------------------
|
257
|
+
ModelCard(
|
258
|
+
model="o4-mini",
|
259
|
+
pricing=Pricing(text=TokenPrice(prompt=1.10, cached_discount=0.275 / 1.1, completion=4.40), audio=None, ft_price_hike=1.5),
|
260
|
+
capabilities=ModelCapabilities(
|
261
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
262
|
+
),
|
263
|
+
temperature_support=False,
|
264
|
+
reasoning_effort_support=True,
|
265
|
+
),
|
266
|
+
ModelCard(
|
267
|
+
model="o4-mini-2025-04-16",
|
268
|
+
pricing=Pricing(text=TokenPrice(prompt=1.10, cached_discount=0.275 / 1.1, completion=4.40), audio=None, ft_price_hike=1.5),
|
269
|
+
capabilities=ModelCapabilities(
|
270
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
271
|
+
),
|
272
|
+
temperature_support=False,
|
273
|
+
reasoning_effort_support=True,
|
274
|
+
),
|
275
|
+
########################
|
276
|
+
########################
|
277
|
+
# ----------------------
|
278
|
+
# Chat models
|
279
|
+
# ----------------------
|
280
|
+
########################
|
281
|
+
########################
|
282
|
+
# ----------------------
|
283
|
+
# gpt-4.1 family
|
284
|
+
# ----------------------
|
285
|
+
ModelCard(
|
286
|
+
model="gpt-4.1",
|
287
|
+
pricing=Pricing(text=TokenPrice(prompt=2.00, cached_discount=0.25, completion=8.00), audio=None, ft_price_hike=1.5),
|
288
|
+
capabilities=ModelCapabilities(
|
289
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
290
|
+
),
|
291
|
+
temperature_support=True,
|
292
|
+
reasoning_effort_support=False,
|
293
|
+
),
|
294
|
+
ModelCard(
|
295
|
+
model="gpt-4.1-2025-04-14",
|
296
|
+
pricing=Pricing(text=TokenPrice(prompt=2.00, cached_discount=0.25, completion=8.00), audio=None, ft_price_hike=1.5),
|
297
|
+
capabilities=ModelCapabilities(
|
298
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
299
|
+
),
|
300
|
+
temperature_support=True,
|
301
|
+
reasoning_effort_support=False,
|
302
|
+
),
|
303
|
+
ModelCard(
|
304
|
+
model="gpt-4.1-mini",
|
305
|
+
pricing=Pricing(text=TokenPrice(prompt=0.40, cached_discount=0.25, completion=1.60), audio=None, ft_price_hike=1.5),
|
306
|
+
capabilities=ModelCapabilities(
|
307
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
308
|
+
),
|
309
|
+
temperature_support=True,
|
310
|
+
reasoning_effort_support=False,
|
311
|
+
),
|
312
|
+
ModelCard(
|
313
|
+
model="gpt-4.1-mini-2025-04-14",
|
314
|
+
pricing=Pricing(text=TokenPrice(prompt=0.40, cached_discount=0.25, completion=1.60), audio=None, ft_price_hike=1.5),
|
315
|
+
capabilities=ModelCapabilities(
|
316
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
317
|
+
),
|
318
|
+
temperature_support=True,
|
319
|
+
reasoning_effort_support=False,
|
320
|
+
),
|
321
|
+
ModelCard(
|
322
|
+
model="gpt-4.1-nano",
|
323
|
+
pricing=Pricing(text=TokenPrice(prompt=0.10, cached_discount=0.25, completion=0.40), audio=None, ft_price_hike=1.5),
|
324
|
+
capabilities=ModelCapabilities(
|
325
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
326
|
+
),
|
327
|
+
temperature_support=True,
|
328
|
+
reasoning_effort_support=False,
|
329
|
+
),
|
330
|
+
ModelCard(
|
331
|
+
model="gpt-4.1-nano-2025-04-14",
|
332
|
+
pricing=Pricing(text=TokenPrice(prompt=0.10, cached_discount=0.25, completion=0.40), audio=None, ft_price_hike=1.5),
|
333
|
+
capabilities=ModelCapabilities(
|
334
|
+
modalities=["text", "image"], endpoints=["chat_completions", "responses", "assistants", "batch"], features=["streaming", "function_calling", "structured_outputs"]
|
335
|
+
),
|
336
|
+
temperature_support=True,
|
337
|
+
reasoning_effort_support=False,
|
338
|
+
),
|
339
|
+
# ----------------------
|
340
|
+
# gpt-4.5 family
|
341
|
+
# ----------------------
|
342
|
+
# ModelCard(
|
343
|
+
# model="gpt-4.5-preview",
|
344
|
+
# pricing=Pricing(
|
345
|
+
# text=TokenPrice(prompt=75, cached_discount=0.5, completion=150.00),
|
346
|
+
# audio=None,
|
347
|
+
# ft_price_hike=1.5
|
348
|
+
# ),
|
349
|
+
# capabilities=ModelCapabilities(
|
350
|
+
# modalities=["text", "image"],
|
351
|
+
# endpoints=["chat_completions", "responses", "assistants", "batch"],
|
352
|
+
# features=["streaming", "function_calling", "structured_outputs"]
|
353
|
+
# ),
|
354
|
+
# logprobs_support=False,
|
355
|
+
# ),
|
356
|
+
# ModelCard(
|
357
|
+
# model="gpt-4.5-preview-2025-02-27",
|
358
|
+
# pricing=Pricing(
|
359
|
+
# text=TokenPrice(prompt=75, cached_discount=0.5, completion=150.00),
|
360
|
+
# audio=None,
|
361
|
+
# ft_price_hike=1.5
|
362
|
+
# ),
|
363
|
+
# capabilities=ModelCapabilities(
|
364
|
+
# modalities=["text", "image"],
|
365
|
+
# endpoints=["chat_completions", "responses", "assistants", "batch"],
|
366
|
+
# features=["streaming", "function_calling", "structured_outputs"]
|
367
|
+
# ),
|
368
|
+
# logprobs_support=False,
|
369
|
+
# ),
|
370
|
+
# ----------------------
|
371
|
+
# gpt-4o family
|
372
|
+
# ----------------------
|
373
|
+
ModelCard(
|
374
|
+
model="chatgpt-4o-latest",
|
375
|
+
pricing=Pricing(text=TokenPrice(prompt=2.50, cached_discount=0.5, completion=10.00), audio=None, ft_price_hike=1.5),
|
376
|
+
capabilities=ModelCapabilities(
|
377
|
+
modalities=["text", "image"],
|
378
|
+
endpoints=["chat_completions", "responses", "assistants", "batch", "fine_tuning"],
|
379
|
+
features=["streaming", "function_calling", "structured_outputs", "fine_tuning", "distillation", "predicted_outputs"],
|
380
|
+
),
|
381
|
+
),
|
382
|
+
ModelCard(
|
383
|
+
model="gpt-4o",
|
384
|
+
pricing=Pricing(text=TokenPrice(prompt=2.50, cached_discount=0.5, completion=10.00), audio=None, ft_price_hike=1.5),
|
385
|
+
capabilities=ModelCapabilities(
|
386
|
+
modalities=["text", "image"],
|
387
|
+
endpoints=["chat_completions", "responses", "assistants", "batch", "fine_tuning"],
|
388
|
+
features=["streaming", "function_calling", "structured_outputs", "fine_tuning", "distillation", "predicted_outputs"],
|
389
|
+
),
|
390
|
+
),
|
391
|
+
ModelCard(
|
392
|
+
model="gpt-4o-2024-08-06",
|
393
|
+
pricing=Pricing(text=TokenPrice(prompt=2.50, cached_discount=0.5, completion=10.00), audio=None, ft_price_hike=1.5),
|
394
|
+
capabilities=ModelCapabilities(
|
395
|
+
modalities=["text", "image"],
|
396
|
+
endpoints=["chat_completions", "responses", "assistants", "batch", "fine_tuning"],
|
397
|
+
features=["streaming", "function_calling", "structured_outputs", "fine_tuning", "distillation", "predicted_outputs"],
|
398
|
+
),
|
399
|
+
),
|
400
|
+
ModelCard(
|
401
|
+
model="gpt-4o-2024-11-20",
|
402
|
+
pricing=Pricing(text=TokenPrice(prompt=2.50, cached_discount=0.5, completion=10.00), audio=None, ft_price_hike=1.5),
|
403
|
+
capabilities=ModelCapabilities(
|
404
|
+
modalities=["text", "image"],
|
405
|
+
endpoints=["chat_completions", "responses", "assistants", "batch", "fine_tuning"],
|
406
|
+
features=["streaming", "function_calling", "structured_outputs", "fine_tuning", "distillation", "predicted_outputs"],
|
407
|
+
),
|
408
|
+
),
|
409
|
+
ModelCard(
|
410
|
+
model="gpt-4o-2024-05-13",
|
411
|
+
pricing=Pricing(text=TokenPrice(prompt=5.00, cached_discount=0.5, completion=15.00), audio=None, ft_price_hike=1.5),
|
412
|
+
capabilities=ModelCapabilities(
|
413
|
+
modalities=["text", "image"],
|
414
|
+
endpoints=["chat_completions", "responses", "assistants", "batch", "fine_tuning"],
|
415
|
+
features=["streaming", "function_calling", "structured_outputs", "fine_tuning", "distillation", "predicted_outputs"],
|
416
|
+
),
|
417
|
+
),
|
418
|
+
# ----------------------
|
419
|
+
# gpt-4o-audio family
|
420
|
+
# ----------------------
|
421
|
+
ModelCard(
|
422
|
+
model="gpt-4o-audio-preview-2024-12-17",
|
423
|
+
pricing=Pricing(text=TokenPrice(prompt=2.50, cached_discount=0.5, completion=10.00), audio=TokenPrice(prompt=40.00, cached_discount=0.2, completion=80.00)),
|
424
|
+
capabilities=ModelCapabilities(modalities=["text", "audio"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
425
|
+
),
|
426
|
+
ModelCard(
|
427
|
+
model="gpt-4o-audio-preview-2024-10-01",
|
428
|
+
pricing=Pricing(text=TokenPrice(prompt=2.50, cached_discount=0.5, completion=10.00), audio=TokenPrice(prompt=100.00, cached_discount=0.2, completion=200.00)),
|
429
|
+
capabilities=ModelCapabilities(modalities=["text", "audio"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
430
|
+
),
|
431
|
+
ModelCard(
|
432
|
+
model="gpt-4o-realtime-preview-2024-12-17",
|
433
|
+
pricing=Pricing(text=TokenPrice(prompt=5.00, cached_discount=0.5, completion=20.00), audio=TokenPrice(prompt=40.00, cached_discount=0.2, completion=80.00)),
|
434
|
+
capabilities=ModelCapabilities(modalities=["text", "audio"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
435
|
+
),
|
436
|
+
ModelCard(
|
437
|
+
model="gpt-4o-realtime-preview-2024-10-01",
|
438
|
+
pricing=Pricing(text=TokenPrice(prompt=5.00, cached_discount=0.5, completion=20.00), audio=TokenPrice(prompt=100.00, cached_discount=0.2, completion=200.00)),
|
439
|
+
capabilities=ModelCapabilities(modalities=["text", "audio"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
440
|
+
),
|
441
|
+
# ----------------------
|
442
|
+
# gpt-4o-mini family
|
443
|
+
# ----------------------
|
444
|
+
ModelCard(
|
445
|
+
model="gpt-4o-mini",
|
446
|
+
pricing=Pricing(text=TokenPrice(prompt=0.15, cached_discount=0.5, completion=0.60), audio=None, ft_price_hike=2.0),
|
447
|
+
capabilities=ModelCapabilities(
|
448
|
+
modalities=["text", "image"],
|
449
|
+
endpoints=["chat_completions", "responses", "assistants", "batch", "fine_tuning"],
|
450
|
+
features=["streaming", "function_calling", "structured_outputs", "fine_tuning"],
|
451
|
+
),
|
452
|
+
),
|
453
|
+
ModelCard(
|
454
|
+
model="gpt-4o-mini-2024-07-18",
|
455
|
+
pricing=Pricing(text=TokenPrice(prompt=0.15, cached_discount=0.5, completion=0.60), audio=None, ft_price_hike=2.0),
|
456
|
+
capabilities=ModelCapabilities(
|
457
|
+
modalities=["text", "image"],
|
458
|
+
endpoints=["chat_completions", "responses", "assistants", "batch", "fine_tuning"],
|
459
|
+
features=["streaming", "function_calling", "structured_outputs", "fine_tuning"],
|
460
|
+
),
|
461
|
+
),
|
462
|
+
# ----------------------
|
463
|
+
# gpt-4o-mini-audio family
|
464
|
+
# ----------------------
|
465
|
+
ModelCard(
|
466
|
+
model="gpt-4o-mini-audio-preview-2024-12-17",
|
467
|
+
pricing=Pricing(
|
468
|
+
text=TokenPrice(prompt=0.15, cached_discount=0.5, completion=0.60), audio=TokenPrice(prompt=10.00, cached_discount=0.2, completion=20.00), ft_price_hike=2.0
|
469
|
+
),
|
470
|
+
capabilities=ModelCapabilities(modalities=["text", "audio"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
471
|
+
),
|
472
|
+
ModelCard(
|
473
|
+
model="gpt-4o-mini-realtime-preview-2024-12-17",
|
474
|
+
pricing=Pricing(
|
475
|
+
text=TokenPrice(prompt=0.60, cached_discount=0.5, completion=2.40), audio=TokenPrice(prompt=10.00, cached_discount=0.2, completion=20.00), ft_price_hike=2.0
|
476
|
+
),
|
477
|
+
capabilities=ModelCapabilities(modalities=["text", "audio"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
478
|
+
),
|
479
|
+
]
|
480
|
+
|
481
|
+
# Anthropic Model Cards
|
482
|
+
anthropic_model_cards = [
|
483
|
+
ModelCard(
|
484
|
+
model="claude-3-5-sonnet-20241022",
|
485
|
+
pricing=Pricing(text=TokenPrice(prompt=3.00, cached_discount=0.5, completion=15.00), audio=None),
|
486
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
487
|
+
),
|
488
|
+
ModelCard(
|
489
|
+
model="claude-3-5-haiku-20241022",
|
490
|
+
pricing=Pricing(text=TokenPrice(prompt=0.80, cached_discount=0.5, completion=4.00), audio=None),
|
491
|
+
capabilities=ModelCapabilities(modalities=["text"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
492
|
+
),
|
493
|
+
ModelCard(
|
494
|
+
model="claude-3-opus-20240229",
|
495
|
+
pricing=Pricing(text=TokenPrice(prompt=15.00, cached_discount=0.5, completion=75.00), audio=None),
|
496
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
497
|
+
),
|
498
|
+
ModelCard(
|
499
|
+
model="claude-3-sonnet-20240229",
|
500
|
+
pricing=Pricing(text=TokenPrice(prompt=3.00, cached_discount=0.5, completion=15.00), audio=None),
|
501
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
502
|
+
),
|
503
|
+
ModelCard(
|
504
|
+
model="claude-3-haiku-20240307",
|
505
|
+
pricing=Pricing(text=TokenPrice(prompt=0.25, cached_discount=0.5, completion=1.25), audio=None),
|
506
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
507
|
+
),
|
508
|
+
]
|
509
|
+
|
510
|
+
# xAI model cards
|
511
|
+
xai_model_cards = [
|
512
|
+
ModelCard(
|
513
|
+
model="grok-2-vision-1212",
|
514
|
+
pricing=Pricing(text=TokenPrice(prompt=2.00, cached_discount=0.5, completion=10.00), audio=None),
|
515
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
516
|
+
),
|
517
|
+
ModelCard(
|
518
|
+
model="grok-2-1212",
|
519
|
+
pricing=Pricing(text=TokenPrice(prompt=2.00, cached_discount=0.5, completion=10.00), audio=None),
|
520
|
+
capabilities=ModelCapabilities(modalities=["text"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
521
|
+
),
|
522
|
+
ModelCard(
|
523
|
+
model="grok-3-beta",
|
524
|
+
pricing=Pricing(text=TokenPrice(prompt=3.00, cached_discount=0.5, completion=15.00), audio=None),
|
525
|
+
capabilities=ModelCapabilities(modalities=["text"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
526
|
+
),
|
527
|
+
ModelCard(
|
528
|
+
model="grok-3-mini-beta",
|
529
|
+
pricing=Pricing(text=TokenPrice(prompt=0.30, cached_discount=0.5, completion=0.50), audio=None),
|
530
|
+
capabilities=ModelCapabilities(modalities=["text"], endpoints=["chat_completions"], features=["streaming", "function_calling"]),
|
531
|
+
),
|
532
|
+
]
|
533
|
+
|
534
|
+
# Add Gemini model cards
|
535
|
+
gemini_model_cards = [
|
536
|
+
# ----------------------
|
537
|
+
# gemini-2.5-pro-exp-03-25 family
|
538
|
+
# ----------------------
|
539
|
+
ModelCard(
|
540
|
+
model="gemini-2.5-pro-exp-03-25",
|
541
|
+
pricing=Pricing(text=TokenPrice(prompt=1.25, cached_discount=0.25, completion=10.00), audio=None),
|
542
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling", "structured_outputs"]),
|
543
|
+
temperature_support=True,
|
544
|
+
),
|
545
|
+
ModelCard(
|
546
|
+
model="gemini-2.5-pro-preview-05-06",
|
547
|
+
pricing=Pricing(text=TokenPrice(prompt=1.25, cached_discount=0.25, completion=10.00), audio=None),
|
548
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling", "structured_outputs"]),
|
549
|
+
),
|
550
|
+
ModelCard(
|
551
|
+
model="gemini-2.5-pro-preview-03-25",
|
552
|
+
pricing=Pricing(text=TokenPrice(prompt=1.25, cached_discount=0.25, completion=10.00), audio=None),
|
553
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling", "structured_outputs"]),
|
554
|
+
temperature_support=True,
|
555
|
+
),
|
556
|
+
ModelCard(
|
557
|
+
model="gemini-2.5-flash-preview-04-17",
|
558
|
+
pricing=Pricing(text=TokenPrice(prompt=0.15, cached_discount=0.25, completion=0.60), audio=None),
|
559
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling", "structured_outputs"]),
|
560
|
+
temperature_support=True,
|
561
|
+
),
|
562
|
+
ModelCard(
|
563
|
+
model="gemini-2.5-flash-preview-05-20",
|
564
|
+
pricing=Pricing(text=TokenPrice(prompt=0.15, cached_discount=0.25, completion=0.60), audio=None),
|
565
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling", "structured_outputs"]),
|
566
|
+
temperature_support=True,
|
567
|
+
),
|
568
|
+
# ----------------------
|
569
|
+
# gemini-2.0-flash family
|
570
|
+
# ----------------------
|
571
|
+
ModelCard(
|
572
|
+
model="gemini-2.0-flash",
|
573
|
+
pricing=Pricing(text=TokenPrice(prompt=0.1, cached_discount=0.025 / 0.1, completion=0.40), audio=TokenPrice(prompt=0.7, cached_discount=0.175 / 0.7, completion=1000)),
|
574
|
+
capabilities=ModelCapabilities(modalities=["text", "image"], endpoints=["chat_completions"], features=["streaming", "function_calling", "structured_outputs"]),
|
575
|
+
temperature_support=True,
|
576
|
+
),
|
577
|
+
ModelCard(
|
578
|
+
model="gemini-2.0-flash-lite",
|
579
|
+
pricing=Pricing(text=TokenPrice(prompt=0.075, cached_discount=1.0, completion=0.30), audio=TokenPrice(prompt=0.075, cached_discount=1.0, completion=1000)),
|
580
|
+
capabilities=ModelCapabilities(
|
581
|
+
modalities=["text", "image", "audio"],
|
582
|
+
endpoints=["chat_completions"],
|
583
|
+
features=["streaming", "structured_outputs"],
|
584
|
+
),
|
585
|
+
temperature_support=True,
|
586
|
+
),
|
587
|
+
]
|
588
|
+
|
589
|
+
|
590
|
+
model_cards = openai_model_cards + gemini_model_cards + xai_model_cards
|
591
|
+
xAI_model_cards = [
|
592
|
+
# ----------------------
|
593
|
+
# grok3-family
|
594
|
+
# ----------------------
|
595
|
+
ModelCard(
|
596
|
+
model="grok-3-beta",
|
597
|
+
pricing=Pricing(text=TokenPrice(prompt=3, cached_discount=1.0, completion=15), audio=None),
|
598
|
+
capabilities=ModelCapabilities(
|
599
|
+
modalities=["text"],
|
600
|
+
endpoints=["chat_completions"],
|
601
|
+
features=["streaming", "structured_outputs"],
|
602
|
+
),
|
603
|
+
temperature_support=True,
|
604
|
+
),
|
605
|
+
ModelCard(
|
606
|
+
model="grok-3-mini-beta",
|
607
|
+
pricing=Pricing(text=TokenPrice(prompt=0.3, cached_discount=1.0, completion=0.5), audio=None),
|
608
|
+
capabilities=ModelCapabilities(
|
609
|
+
modalities=["text"],
|
610
|
+
endpoints=["chat_completions"],
|
611
|
+
features=["streaming", "structured_outputs"],
|
612
|
+
),
|
613
|
+
temperature_support=True,
|
614
|
+
),
|
615
|
+
]
|
616
|
+
|
617
|
+
|
618
|
+
model_cards = openai_model_cards + gemini_model_cards + xAI_model_cards
|
619
|
+
|
620
|
+
|
621
|
+
def get_model_card(model: str) -> ModelCard:
|
622
|
+
"""
|
623
|
+
Get the model card for a specific model.
|
624
|
+
|
625
|
+
Args:
|
626
|
+
model: The model name to look up
|
627
|
+
|
628
|
+
Returns:
|
629
|
+
The ModelCard for the specified model
|
630
|
+
|
631
|
+
Raises:
|
632
|
+
ValueError: If no model card is found for the specified model
|
633
|
+
"""
|
634
|
+
# Extract base model name for fine-tuned models like "ft:gpt-4o:uiform:4389573"
|
635
|
+
if model.startswith("ft:"):
|
636
|
+
# Split by colon and take the second part (index 1) which contains the base model
|
637
|
+
parts = model.split(":")
|
638
|
+
if len(parts) > 1:
|
639
|
+
model = parts[1]
|
640
|
+
|
641
|
+
for card in model_cards:
|
642
|
+
if card.model == model:
|
643
|
+
return card
|
644
|
+
|
645
|
+
raise ValueError(f"No model card found for model: {model}")
|
File without changes
|