File size: 6,064 Bytes
7db0ae4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
#### What this tests ####
#    This tests calling batch_completions by running 100 messages together

import sys, os, time
import traceback, asyncio
import pytest

sys.path.insert(
    0, os.path.abspath("../..")
)  # Adds the parent directory to the system path
import litellm
from litellm import Router
import concurrent
from dotenv import load_dotenv

load_dotenv()

model_list = [
    {  # list of model deployments
        "model_name": "gpt-3.5-turbo",  # openai model name
        "litellm_params": {  # params for litellm completion/embedding call
            "model": "azure/chatgpt-v-2",
            "api_key": "bad-key",
            "api_version": os.getenv("AZURE_API_VERSION"),
            "api_base": os.getenv("AZURE_API_BASE"),
        },
        "tpm": 240000,
        "rpm": 1800,
    },
    {
        "model_name": "gpt-3.5-turbo",  # openai model name
        "litellm_params": {  # params for litellm completion/embedding call
            "model": "gpt-3.5-turbo",
            "api_key": os.getenv("OPENAI_API_KEY"),
        },
        "tpm": 1000000,
        "rpm": 9000,
    },
]

kwargs = {
    "model": "gpt-3.5-turbo",
    "messages": [{"role": "user", "content": "Hey, how's it going?"}],
}


def test_multiple_deployments_sync():
    import concurrent, time

    litellm.set_verbose = False
    results = []
    router = Router(
        model_list=model_list,
        redis_host=os.getenv("REDIS_HOST"),
        redis_password=os.getenv("REDIS_PASSWORD"),
        redis_port=int(os.getenv("REDIS_PORT")),  # type: ignore
        routing_strategy="simple-shuffle",
        set_verbose=True,
        num_retries=1,
    )  # type: ignore
    try:
        for _ in range(3):
            response = router.completion(**kwargs)
            results.append(response)
        print(results)
        router.reset()
    except Exception as e:
        print(f"FAILED TEST!")
        pytest.fail(f"An error occurred - {traceback.format_exc()}")


# test_multiple_deployments_sync()


def test_multiple_deployments_parallel():
    litellm.set_verbose = False  # Corrected the syntax for setting verbose to False
    results = []
    futures = {}
    start_time = time.time()
    router = Router(
        model_list=model_list,
        redis_host=os.getenv("REDIS_HOST"),
        redis_password=os.getenv("REDIS_PASSWORD"),
        redis_port=int(os.getenv("REDIS_PORT")),  # type: ignore
        routing_strategy="simple-shuffle",
        set_verbose=True,
        num_retries=1,
    )  # type: ignore
    # Assuming you have an executor instance defined somewhere in your code
    with concurrent.futures.ThreadPoolExecutor() as executor:
        for _ in range(5):
            future = executor.submit(router.completion, **kwargs)
            futures[future] = future

        # Retrieve the results from the futures
        while futures:
            done, not_done = concurrent.futures.wait(
                futures.values(),
                timeout=10,
                return_when=concurrent.futures.FIRST_COMPLETED,
            )
            for future in done:
                try:
                    result = future.result()
                    results.append(result)
                    del futures[future]  # Remove the done future
                except Exception as e:
                    print(f"Exception: {e}; traceback: {traceback.format_exc()}")
                    del futures[future]  # Remove the done future with exception

            print(f"Remaining futures: {len(futures)}")
    router.reset()
    end_time = time.time()
    print(results)
    print(f"ELAPSED TIME: {end_time - start_time}")


# Assuming litellm, router, and executor are defined somewhere in your code


# test_multiple_deployments_parallel()
def test_cooldown_same_model_name():
    # users could have the same model with different api_base
    # example
    # azure/chatgpt, api_base: 1234
    # azure/chatgpt, api_base: 1235
    # if 1234 fails, it should only cooldown 1234 and then try with 1235
    litellm.set_verbose = False
    try:
        print("testing cooldown same model name")
        model_list = [
            {
                "model_name": "gpt-3.5-turbo",
                "litellm_params": {
                    "model": "azure/chatgpt-v-2",
                    "api_key": os.getenv("AZURE_API_KEY"),
                    "api_version": os.getenv("AZURE_API_VERSION"),
                    "api_base": "BAD_API_BASE",
                    "tpm": 90,
                },
            },
            {
                "model_name": "gpt-3.5-turbo",
                "litellm_params": {
                    "model": "azure/chatgpt-v-2",
                    "api_key": os.getenv("AZURE_API_KEY"),
                    "api_version": os.getenv("AZURE_API_VERSION"),
                    "api_base": os.getenv("AZURE_API_BASE"),
                    "tpm": 0.000001,
                },
            },
        ]

        router = Router(
            model_list=model_list,
            redis_host=os.getenv("REDIS_HOST"),
            redis_password=os.getenv("REDIS_PASSWORD"),
            redis_port=int(os.getenv("REDIS_PORT")),
            routing_strategy="simple-shuffle",
            set_verbose=True,
            num_retries=3,
        )  # type: ignore

        response = router.completion(
            model="gpt-3.5-turbo",
            messages=[{"role": "user", "content": "hello this request will pass"}],
        )
        print(router.model_list)
        model_ids = []
        for model in router.model_list:
            model_ids.append(model["model_info"]["id"])
        print("\n litellm model ids ", model_ids)

        # example litellm_model_names ['azure/chatgpt-v-2-ModelID-64321', 'azure/chatgpt-v-2-ModelID-63960']
        assert (
            model_ids[0] != model_ids[1]
        )  # ensure both models have a uuid added, and they have different names

        print("\ngot response\n", response)
    except Exception as e:
        pytest.fail(f"Got unexpected exception on router! - {e}")


test_cooldown_same_model_name()