llmstudio-core 0.0.1a1__tar.gz
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.
- llmstudio_core-0.0.1a1/PKG-INFO +79 -0
- llmstudio_core-0.0.1a1/README.md +57 -0
- llmstudio_core-0.0.1a1/llmstudio_core/__init__.py +0 -0
- llmstudio_core-0.0.1a1/llmstudio_core/config.yaml +374 -0
- llmstudio_core-0.0.1a1/llmstudio_core/exceptions.py +6 -0
- llmstudio_core-0.0.1a1/llmstudio_core/providers/__init__.py +35 -0
- llmstudio_core-0.0.1a1/llmstudio_core/providers/azure.py +601 -0
- llmstudio_core-0.0.1a1/llmstudio_core/providers/openai.py +70 -0
- llmstudio_core-0.0.1a1/llmstudio_core/providers/provider.py +676 -0
- llmstudio_core-0.0.1a1/llmstudio_core/providers/vertexai.py +569 -0
- llmstudio_core-0.0.1a1/llmstudio_core/utils.py +65 -0
- llmstudio_core-0.0.1a1/pyproject.toml +29 -0
|
@@ -0,0 +1,79 @@
|
|
|
1
|
+
Metadata-Version: 2.1
|
|
2
|
+
Name: llmstudio-core
|
|
3
|
+
Version: 0.0.1a1
|
|
4
|
+
Summary: LLMStudio core capabilities for routing llm calls for any vendor. No proxy server required. For that use llmstudio[proxy]
|
|
5
|
+
License: MIT
|
|
6
|
+
Author: ClΓ‘udio Lemos
|
|
7
|
+
Author-email: claudio.lemos@tensorops.ai
|
|
8
|
+
Requires-Python: >=3.9,<4.0
|
|
9
|
+
Classifier: License :: OSI Approved :: MIT License
|
|
10
|
+
Classifier: Programming Language :: Python :: 3
|
|
11
|
+
Classifier: Programming Language :: Python :: 3.9
|
|
12
|
+
Classifier: Programming Language :: Python :: 3.10
|
|
13
|
+
Classifier: Programming Language :: Python :: 3.11
|
|
14
|
+
Classifier: Programming Language :: Python :: 3.12
|
|
15
|
+
Requires-Dist: openai (>=1.26,<2.0)
|
|
16
|
+
Requires-Dist: pydantic (>=2.5,<3.0)
|
|
17
|
+
Requires-Dist: pyyaml (>=6,<7)
|
|
18
|
+
Requires-Dist: requests (>=2.31,<3.0)
|
|
19
|
+
Requires-Dist: tiktoken (>=0.7,<0.8)
|
|
20
|
+
Description-Content-Type: text/markdown
|
|
21
|
+
|
|
22
|
+
# LLMstudio-core by [TensorOps](http://tensorops.ai "TensorOps")
|
|
23
|
+
|
|
24
|
+
Prompt Engineering at your fingertips
|
|
25
|
+
|
|
26
|
+
|
|
27
|
+
|
|
28
|
+
## π Features
|
|
29
|
+
- **Custom and Local LLM Support**: Use custom or local open-source LLMs through Ollama.
|
|
30
|
+
- **Python SDK**: Easily integrate LLMstudio into your existing workflows.
|
|
31
|
+
- **LangChain Integration**: LLMstudio integrates with your already existing LangChain projects.
|
|
32
|
+
|
|
33
|
+
## π Quickstart
|
|
34
|
+
|
|
35
|
+
Don't forget to check out [https://docs.llmstudio.ai](docs) page.
|
|
36
|
+
|
|
37
|
+
## Installation
|
|
38
|
+
|
|
39
|
+
Install the latest version of **LLMstudio** using `pip`. We suggest that you create and activate a new environment using `conda`
|
|
40
|
+
|
|
41
|
+
```bash
|
|
42
|
+
pip install llmstudio-core
|
|
43
|
+
```
|
|
44
|
+
|
|
45
|
+
Create a `.env` file at the same path you'll run **LLMstudio**
|
|
46
|
+
|
|
47
|
+
```bash
|
|
48
|
+
OPENAI_API_KEY="sk-api_key"
|
|
49
|
+
GOOGLE_API_KEY="sk-api_key"
|
|
50
|
+
```
|
|
51
|
+
|
|
52
|
+
Now you should be able to run **LLMstudio** Providers using the following code:
|
|
53
|
+
|
|
54
|
+
```
|
|
55
|
+
# You can set OPENAI_API_KEY environment variable, add it to .env, or pass directly as api_key
|
|
56
|
+
import os
|
|
57
|
+
from llmstudio_core.providers import LLMCore as LLM
|
|
58
|
+
llm = LLM("vertexai", api_key=os.environ["GOOGLE_API_KEY"])
|
|
59
|
+
response = llm.chat("How are you", model="gemini-1.5-pro-latest")
|
|
60
|
+
print(response.chat_output, response.metrics)
|
|
61
|
+
```
|
|
62
|
+
## π Documentation
|
|
63
|
+
|
|
64
|
+
- [Visit our docs to learn how the SDK works](https://docs.LLMstudio.ai) (coming soon)
|
|
65
|
+
- Checkout our [notebook examples](https://github.com/TensorOpsAI/LLMstudio/tree/main/examples) to follow along with interactive tutorials
|
|
66
|
+
|
|
67
|
+
## π¨βπ» Contributing
|
|
68
|
+
|
|
69
|
+
- Head on to our [Contribution Guide](https://github.com/TensorOpsAI/LLMstudio/tree/main/CONTRIBUTING.md) to see how you can help LLMstudio.
|
|
70
|
+
- Join our [Discord](https://discord.gg/GkAfPZR9wy) to talk with other LLMstudio enthusiasts.
|
|
71
|
+
|
|
72
|
+
## Training
|
|
73
|
+
|
|
74
|
+
[](https://www.tensorops.ai/llm-studio-workshop)
|
|
75
|
+
|
|
76
|
+
---
|
|
77
|
+
|
|
78
|
+
Thank you for choosing LLMstudio. Your journey to perfecting AI interactions starts here.
|
|
79
|
+
|
|
@@ -0,0 +1,57 @@
|
|
|
1
|
+
# LLMstudio-core by [TensorOps](http://tensorops.ai "TensorOps")
|
|
2
|
+
|
|
3
|
+
Prompt Engineering at your fingertips
|
|
4
|
+
|
|
5
|
+
|
|
6
|
+
|
|
7
|
+
## π Features
|
|
8
|
+
- **Custom and Local LLM Support**: Use custom or local open-source LLMs through Ollama.
|
|
9
|
+
- **Python SDK**: Easily integrate LLMstudio into your existing workflows.
|
|
10
|
+
- **LangChain Integration**: LLMstudio integrates with your already existing LangChain projects.
|
|
11
|
+
|
|
12
|
+
## π Quickstart
|
|
13
|
+
|
|
14
|
+
Don't forget to check out [https://docs.llmstudio.ai](docs) page.
|
|
15
|
+
|
|
16
|
+
## Installation
|
|
17
|
+
|
|
18
|
+
Install the latest version of **LLMstudio** using `pip`. We suggest that you create and activate a new environment using `conda`
|
|
19
|
+
|
|
20
|
+
```bash
|
|
21
|
+
pip install llmstudio-core
|
|
22
|
+
```
|
|
23
|
+
|
|
24
|
+
Create a `.env` file at the same path you'll run **LLMstudio**
|
|
25
|
+
|
|
26
|
+
```bash
|
|
27
|
+
OPENAI_API_KEY="sk-api_key"
|
|
28
|
+
GOOGLE_API_KEY="sk-api_key"
|
|
29
|
+
```
|
|
30
|
+
|
|
31
|
+
Now you should be able to run **LLMstudio** Providers using the following code:
|
|
32
|
+
|
|
33
|
+
```
|
|
34
|
+
# You can set OPENAI_API_KEY environment variable, add it to .env, or pass directly as api_key
|
|
35
|
+
import os
|
|
36
|
+
from llmstudio_core.providers import LLMCore as LLM
|
|
37
|
+
llm = LLM("vertexai", api_key=os.environ["GOOGLE_API_KEY"])
|
|
38
|
+
response = llm.chat("How are you", model="gemini-1.5-pro-latest")
|
|
39
|
+
print(response.chat_output, response.metrics)
|
|
40
|
+
```
|
|
41
|
+
## π Documentation
|
|
42
|
+
|
|
43
|
+
- [Visit our docs to learn how the SDK works](https://docs.LLMstudio.ai) (coming soon)
|
|
44
|
+
- Checkout our [notebook examples](https://github.com/TensorOpsAI/LLMstudio/tree/main/examples) to follow along with interactive tutorials
|
|
45
|
+
|
|
46
|
+
## π¨βπ» Contributing
|
|
47
|
+
|
|
48
|
+
- Head on to our [Contribution Guide](https://github.com/TensorOpsAI/LLMstudio/tree/main/CONTRIBUTING.md) to see how you can help LLMstudio.
|
|
49
|
+
- Join our [Discord](https://discord.gg/GkAfPZR9wy) to talk with other LLMstudio enthusiasts.
|
|
50
|
+
|
|
51
|
+
## Training
|
|
52
|
+
|
|
53
|
+
[](https://www.tensorops.ai/llm-studio-workshop)
|
|
54
|
+
|
|
55
|
+
---
|
|
56
|
+
|
|
57
|
+
Thank you for choosing LLMstudio. Your journey to perfecting AI interactions starts here.
|
|
File without changes
|
|
@@ -0,0 +1,374 @@
|
|
|
1
|
+
providers:
|
|
2
|
+
anthropic:
|
|
3
|
+
id: anthropic
|
|
4
|
+
name: Anthropic
|
|
5
|
+
chat: true
|
|
6
|
+
embed: true
|
|
7
|
+
keys:
|
|
8
|
+
- ANTHROPIC_API_KEY
|
|
9
|
+
models:
|
|
10
|
+
claude-3-opus-20240229:
|
|
11
|
+
mode: chat
|
|
12
|
+
max_tokens: 200000
|
|
13
|
+
input_token_cost: 0.000015
|
|
14
|
+
output_token_cost: 0.000075
|
|
15
|
+
claude-3-sonnet-2024022:
|
|
16
|
+
mode: chat
|
|
17
|
+
max_tokens: 200000
|
|
18
|
+
input_token_cost: 0.000003
|
|
19
|
+
output_token_cost: 0.000015
|
|
20
|
+
claude-3-haiku-20240307:
|
|
21
|
+
mode: chat
|
|
22
|
+
max_tokens: 200000
|
|
23
|
+
input_token_cost: 0.00000025
|
|
24
|
+
output_token_cost: 0.00000125
|
|
25
|
+
claude-2.1:
|
|
26
|
+
mode: chat
|
|
27
|
+
max_tokens: 200000
|
|
28
|
+
input_token_cost: 0.000008
|
|
29
|
+
output_token_cost: 0.000024
|
|
30
|
+
claude-2:
|
|
31
|
+
mode: chat
|
|
32
|
+
max_tokens: 100000
|
|
33
|
+
input_token_cost: 0.000008
|
|
34
|
+
output_token_cost: 0.000024
|
|
35
|
+
claude-instant-1.2:
|
|
36
|
+
mode: chat
|
|
37
|
+
max_tokens: 100000
|
|
38
|
+
input_token_cost: 0.00000163
|
|
39
|
+
output_token_cost: 0.00000551
|
|
40
|
+
parameters:
|
|
41
|
+
temperature:
|
|
42
|
+
name: "Temperature"
|
|
43
|
+
type: float
|
|
44
|
+
default: 1
|
|
45
|
+
min: 0
|
|
46
|
+
max: 1
|
|
47
|
+
step: 0.01
|
|
48
|
+
max_tokens:
|
|
49
|
+
name: "Maximum tokens"
|
|
50
|
+
type: float
|
|
51
|
+
default: 256
|
|
52
|
+
min: 1
|
|
53
|
+
max: 4096
|
|
54
|
+
step: 0.01
|
|
55
|
+
top_p:
|
|
56
|
+
name: "Top P"
|
|
57
|
+
type: float
|
|
58
|
+
default: 1
|
|
59
|
+
min: 0
|
|
60
|
+
max: 1
|
|
61
|
+
step: 0.01
|
|
62
|
+
top_k:
|
|
63
|
+
name: "Top K"
|
|
64
|
+
type: float
|
|
65
|
+
default: 5
|
|
66
|
+
min: 0
|
|
67
|
+
max: 500
|
|
68
|
+
step: 1
|
|
69
|
+
ollama:
|
|
70
|
+
id: ollama
|
|
71
|
+
name: Ollama
|
|
72
|
+
chat: true
|
|
73
|
+
embed: true
|
|
74
|
+
keys:
|
|
75
|
+
models:
|
|
76
|
+
llama2:
|
|
77
|
+
mode: chat
|
|
78
|
+
max_tokens: 0
|
|
79
|
+
input_token_cost: 0
|
|
80
|
+
output_token_cost: 0
|
|
81
|
+
parameters:
|
|
82
|
+
temperature:
|
|
83
|
+
name: "Temperature"
|
|
84
|
+
type: float
|
|
85
|
+
default: 1
|
|
86
|
+
min: 0
|
|
87
|
+
max: 1
|
|
88
|
+
step: 0.01
|
|
89
|
+
max_tokens:
|
|
90
|
+
name: "Maximum tokens"
|
|
91
|
+
type: float
|
|
92
|
+
default: 256
|
|
93
|
+
min: 1
|
|
94
|
+
max: 4096
|
|
95
|
+
step: 0.01
|
|
96
|
+
top_p:
|
|
97
|
+
name: "Top P"
|
|
98
|
+
type: float
|
|
99
|
+
default: 1
|
|
100
|
+
min: 0
|
|
101
|
+
max: 1
|
|
102
|
+
step: 0.01
|
|
103
|
+
top_k:
|
|
104
|
+
name: "Top K"
|
|
105
|
+
type: float
|
|
106
|
+
default: 5
|
|
107
|
+
min: 0
|
|
108
|
+
max: 500
|
|
109
|
+
step: 1
|
|
110
|
+
openai:
|
|
111
|
+
id: openai
|
|
112
|
+
name: OpenAI
|
|
113
|
+
chat: true
|
|
114
|
+
embed: true
|
|
115
|
+
keys:
|
|
116
|
+
- OPENAI_API_KEY
|
|
117
|
+
models:
|
|
118
|
+
gpt-4o-mini:
|
|
119
|
+
mode: chat
|
|
120
|
+
max_tokens: 128000
|
|
121
|
+
input_token_cost: 0.00000015
|
|
122
|
+
output_token_cost: 0.00000060
|
|
123
|
+
gpt-4o:
|
|
124
|
+
mode: chat
|
|
125
|
+
max_tokens: 128000
|
|
126
|
+
input_token_cost: 0.000005
|
|
127
|
+
output_token_cost: 0.000015
|
|
128
|
+
gpt-4-turbo:
|
|
129
|
+
mode: chat
|
|
130
|
+
max_tokens: 128000
|
|
131
|
+
input_token_cost: 0.00001
|
|
132
|
+
output_token_cost: 0.00003
|
|
133
|
+
gpt-4:
|
|
134
|
+
mode: chat
|
|
135
|
+
max_tokens: 8192
|
|
136
|
+
input_token_cost: 0.00003
|
|
137
|
+
output_token_cost: 0.00006
|
|
138
|
+
gpt-3.5-turbo:
|
|
139
|
+
mode: chat
|
|
140
|
+
max_tokens: 16385
|
|
141
|
+
input_token_cost: 0.0000005
|
|
142
|
+
output_token_cost: 0.0000015
|
|
143
|
+
gpt-3.5-turbo-instruct:
|
|
144
|
+
mode: chat
|
|
145
|
+
max_tokens: 4096
|
|
146
|
+
input_token_cost: 0.0000015
|
|
147
|
+
output_token_cost: 0.000002
|
|
148
|
+
parameters:
|
|
149
|
+
temperature:
|
|
150
|
+
name: "Temperature"
|
|
151
|
+
type: float
|
|
152
|
+
default: 1
|
|
153
|
+
min: 0
|
|
154
|
+
max: 2
|
|
155
|
+
step: 0.01
|
|
156
|
+
max_tokens:
|
|
157
|
+
name: "Maximum length"
|
|
158
|
+
type: int
|
|
159
|
+
default: 256
|
|
160
|
+
min: 1
|
|
161
|
+
max: 4096
|
|
162
|
+
step: 1
|
|
163
|
+
top_p:
|
|
164
|
+
name: "Top P"
|
|
165
|
+
type: float
|
|
166
|
+
default: 1
|
|
167
|
+
min: 0
|
|
168
|
+
max: 1
|
|
169
|
+
step: 0.01
|
|
170
|
+
frequency_penalty:
|
|
171
|
+
name: "Frequency Penalty"
|
|
172
|
+
type: float
|
|
173
|
+
default: 0
|
|
174
|
+
min: 0
|
|
175
|
+
max: 2
|
|
176
|
+
step: 0.01
|
|
177
|
+
presence_penalty:
|
|
178
|
+
name: "Presence Penalty"
|
|
179
|
+
type: float
|
|
180
|
+
default: 0
|
|
181
|
+
min: 0
|
|
182
|
+
max: 2
|
|
183
|
+
step: 0.01
|
|
184
|
+
azure:
|
|
185
|
+
id: azure
|
|
186
|
+
name: Azure
|
|
187
|
+
chat: true
|
|
188
|
+
embed: true
|
|
189
|
+
keys:
|
|
190
|
+
- AZURE_API_KEY
|
|
191
|
+
- AZURE_API_ENDPOINT
|
|
192
|
+
- AZURE_API_VERSION
|
|
193
|
+
models:
|
|
194
|
+
gpt-4o-mini:
|
|
195
|
+
mode: chat
|
|
196
|
+
max_tokens: 128000
|
|
197
|
+
input_token_cost: 0.00000015
|
|
198
|
+
output_token_cost: 0.0000006
|
|
199
|
+
gpt-4o:
|
|
200
|
+
mode: chat
|
|
201
|
+
max_tokens: 128000
|
|
202
|
+
input_token_cost: 0.000005
|
|
203
|
+
output_token_cost: 0.000015
|
|
204
|
+
gpt-4-turbo:
|
|
205
|
+
mode: chat
|
|
206
|
+
max_tokens: 128000
|
|
207
|
+
input_token_cost: 0.00001
|
|
208
|
+
output_token_cost: 0.00003
|
|
209
|
+
gpt-4:
|
|
210
|
+
mode: chat
|
|
211
|
+
max_tokens: 8192
|
|
212
|
+
input_token_cost: 0.00003
|
|
213
|
+
output_token_cost: 0.00006
|
|
214
|
+
gpt-35-turbo:
|
|
215
|
+
mode: chat
|
|
216
|
+
max_tokens: 4097
|
|
217
|
+
input_token_cost: 0.0000005
|
|
218
|
+
output_token_cost: 0.0000015
|
|
219
|
+
Meta-Llama-3.1-405B-Instruct:
|
|
220
|
+
mode: chat
|
|
221
|
+
max_tokens: 128000
|
|
222
|
+
input_token_cost: 0.00533
|
|
223
|
+
output_token_cost: 0.016
|
|
224
|
+
Meta-Llama-3.1-70B-Instruct:
|
|
225
|
+
mode: chat
|
|
226
|
+
max_tokens: 128000
|
|
227
|
+
input_token_cost: 0.00268
|
|
228
|
+
output_token_cost: 0.00354
|
|
229
|
+
Meta-Llama-3.1-8B-Instruct:
|
|
230
|
+
mode: chat
|
|
231
|
+
max_tokens: 128000
|
|
232
|
+
input_token_cost: 0.0003
|
|
233
|
+
output_token_cost: 0.00061
|
|
234
|
+
Meta-Llama-3-70B-Instruct:
|
|
235
|
+
mode: chat
|
|
236
|
+
max_tokens: 8192
|
|
237
|
+
input_token_cost: 0.00378
|
|
238
|
+
output_token_cost: 0.01134
|
|
239
|
+
Meta-Llama-3-8B-Instruct:
|
|
240
|
+
mode: chat
|
|
241
|
+
max_tokens: 8192
|
|
242
|
+
input_token_cost: 0.00037
|
|
243
|
+
output_token_cost: 0.0011
|
|
244
|
+
Meta-Llama-2-70B:
|
|
245
|
+
mode: chat
|
|
246
|
+
max_tokens: 4096
|
|
247
|
+
input_token_cost: 0.00154
|
|
248
|
+
output_token_cost: 0.00177
|
|
249
|
+
Meta-Llama-2-70B-Chat:
|
|
250
|
+
mode: chat
|
|
251
|
+
max_tokens: 4096
|
|
252
|
+
input_token_cost: 0.00154
|
|
253
|
+
output_token_cost: 0.00177
|
|
254
|
+
Meta-Llama-2-13B-Chat:
|
|
255
|
+
mode: chat
|
|
256
|
+
max_tokens: 4096
|
|
257
|
+
input_token_cost: 0.00081
|
|
258
|
+
output_token_cost: 0.00094
|
|
259
|
+
Meta-Llama-2-7B:
|
|
260
|
+
mode: chat
|
|
261
|
+
max_tokens: 4096
|
|
262
|
+
input_token_cost: 0.00052
|
|
263
|
+
output_token_cost: 0.00067
|
|
264
|
+
parameters:
|
|
265
|
+
temperature:
|
|
266
|
+
name: "Temperature"
|
|
267
|
+
type: float
|
|
268
|
+
default: 1
|
|
269
|
+
min: 0
|
|
270
|
+
max: 2
|
|
271
|
+
step: 0.01
|
|
272
|
+
max_tokens:
|
|
273
|
+
name: "Maximum length"
|
|
274
|
+
type: int
|
|
275
|
+
default: 256
|
|
276
|
+
min: 1
|
|
277
|
+
max: 4096
|
|
278
|
+
step: 1
|
|
279
|
+
top_p:
|
|
280
|
+
name: "Top P"
|
|
281
|
+
type: float
|
|
282
|
+
default: 1
|
|
283
|
+
min: 0
|
|
284
|
+
max: 1
|
|
285
|
+
step: 0.01
|
|
286
|
+
frequency_penalty:
|
|
287
|
+
name: "Frequency Penalty"
|
|
288
|
+
type: float
|
|
289
|
+
default: 0
|
|
290
|
+
min: 0
|
|
291
|
+
max: 2
|
|
292
|
+
step: 0.01
|
|
293
|
+
presence_penalty:
|
|
294
|
+
name: "Presence Penalty"
|
|
295
|
+
type: float
|
|
296
|
+
default: 0
|
|
297
|
+
min: 0
|
|
298
|
+
max: 2
|
|
299
|
+
step: 0.01
|
|
300
|
+
vertexai:
|
|
301
|
+
id: vertexai
|
|
302
|
+
name: VertexAI
|
|
303
|
+
chat: true
|
|
304
|
+
embed: true
|
|
305
|
+
keys:
|
|
306
|
+
- GOOGLE_API_KEY
|
|
307
|
+
models:
|
|
308
|
+
gemini-1.5-flash:
|
|
309
|
+
mode: chat
|
|
310
|
+
max_tokens: 1000000
|
|
311
|
+
input_token_cost:
|
|
312
|
+
- range: [0, 128000]
|
|
313
|
+
cost: 0.00000035
|
|
314
|
+
- range: [128001, null]
|
|
315
|
+
cost: 0.0000007
|
|
316
|
+
output_token_cost:
|
|
317
|
+
- range: [0, 128000]
|
|
318
|
+
cost: 0.00000105
|
|
319
|
+
- range: [128001, null]
|
|
320
|
+
cost: 0.0000021
|
|
321
|
+
gemini-1.5-pro-latest:
|
|
322
|
+
mode: chat
|
|
323
|
+
max_tokens: 1000000
|
|
324
|
+
input_token_cost:
|
|
325
|
+
- range: [0, 128000]
|
|
326
|
+
cost: 0.00000035
|
|
327
|
+
- range: [128001, null]
|
|
328
|
+
cost: 0.0000007
|
|
329
|
+
output_token_cost:
|
|
330
|
+
- range: [0, 128000]
|
|
331
|
+
cost: 0.0000105
|
|
332
|
+
- range: [128001, null]
|
|
333
|
+
cost: 0.000021
|
|
334
|
+
gemini-1.0-pro:
|
|
335
|
+
mode: chat
|
|
336
|
+
max_tokens: 1000000
|
|
337
|
+
input_token_cost: 0.0000005
|
|
338
|
+
output_token_cost: 0.0000015
|
|
339
|
+
parameters:
|
|
340
|
+
temperature:
|
|
341
|
+
name: "Temperature"
|
|
342
|
+
type: float
|
|
343
|
+
default: 1
|
|
344
|
+
min: 0
|
|
345
|
+
max: 2
|
|
346
|
+
step: 0.01
|
|
347
|
+
max_tokens:
|
|
348
|
+
name: "Maximum length"
|
|
349
|
+
type: int
|
|
350
|
+
default: 256
|
|
351
|
+
min: 1
|
|
352
|
+
max: 4096
|
|
353
|
+
step: 1
|
|
354
|
+
top_p:
|
|
355
|
+
name: "Top P"
|
|
356
|
+
type: float
|
|
357
|
+
default: 1
|
|
358
|
+
min: 0
|
|
359
|
+
max: 1
|
|
360
|
+
step: 0.01
|
|
361
|
+
frequency_penalty:
|
|
362
|
+
name: "Frequency Penalty"
|
|
363
|
+
type: float
|
|
364
|
+
default: 0
|
|
365
|
+
min: 0
|
|
366
|
+
max: 2
|
|
367
|
+
step: 0.01
|
|
368
|
+
presence_penalty:
|
|
369
|
+
name: "Presence Penalty"
|
|
370
|
+
type: float
|
|
371
|
+
default: 0
|
|
372
|
+
min: 0
|
|
373
|
+
max: 2
|
|
374
|
+
step: 0.01
|
|
@@ -0,0 +1,35 @@
|
|
|
1
|
+
# from llmstudio_core.providers.anthropic import AnthropicProvider #TODO: adpat it
|
|
2
|
+
from typing import Optional
|
|
3
|
+
|
|
4
|
+
from llmstudio_core.providers.azure import AzureProvider
|
|
5
|
+
|
|
6
|
+
# from llmstudio_core.providers.ollama import OllamaProvider #TODO: adapt it
|
|
7
|
+
from llmstudio_core.providers.openai import OpenAIProvider
|
|
8
|
+
from llmstudio_core.providers.provider import ProviderCore, provider_registry
|
|
9
|
+
from llmstudio_core.providers.vertexai import VertexAIProvider
|
|
10
|
+
from llmstudio_core.utils import _load_providers_config
|
|
11
|
+
|
|
12
|
+
_engine_config = _load_providers_config()
|
|
13
|
+
|
|
14
|
+
|
|
15
|
+
def LLMCore(provider: str, api_key: Optional[str] = None, **kwargs) -> ProviderCore:
|
|
16
|
+
"""
|
|
17
|
+
Factory method to create an instance of a provider.
|
|
18
|
+
|
|
19
|
+
Args:
|
|
20
|
+
provider (str): The name of the provider.
|
|
21
|
+
api_key (Optional[str], optional): The API key for the provider. Defaults to None.
|
|
22
|
+
|
|
23
|
+
Returns:
|
|
24
|
+
ProviderCore: An instance of the provider.
|
|
25
|
+
|
|
26
|
+
Raises:
|
|
27
|
+
NotImplementedError: If the provider is not found in the provider map.
|
|
28
|
+
"""
|
|
29
|
+
provider_config = _engine_config.providers.get(provider)
|
|
30
|
+
provider_class = provider_registry.get(provider_config.id)
|
|
31
|
+
if provider_class:
|
|
32
|
+
return provider_class(config=provider_config, api_key=api_key, **kwargs)
|
|
33
|
+
raise NotImplementedError(
|
|
34
|
+
f"Provider not found: {provider_config.id}. Available providers: {str(provider_registry.keys())}"
|
|
35
|
+
)
|