llmsim 0.4.0

LLM Traffic Simulator - A lightweight, high-performance LLM API simulator
Documentation 
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

#!/usr/bin/env -S uv run --script
# /// script
# requires-python = ">=3.10"
# dependencies = [
#     "langchain-openai>=0.3.0",
# ]
# ///
"""
LangChain client example for llmsim.

This script demonstrates connecting to a running llmsim server using LangChain's
OpenAI-compatible interface. The server simulates LLM responses with realistic
latency without running actual models.

Server endpoints:
    POST /openai/v1/chat/completions - Chat completions (streaming supported)
    POST /openai/v1/responses        - Responses API
    GET  /openai/v1/models           - List available models
    GET  /openai/v1/models/:id       - Get model details

Prerequisites:
    Start the llmsim server first:
        llmsim serve --port 8080

    Or from source:
        cargo run --release -- serve --port 8080

Usage:
    uv run examples/langchain_client.py

Environment variables:
    LLMSIM_URL: Server URL (default: http://localhost:8080/openai/v1)
"""

import os
import sys

from langchain_openai import ChatOpenAI
from langchain_core.messages import HumanMessage, SystemMessage


def main() -> None:
    base_url = os.environ.get("LLMSIM_URL", "http://localhost:8080/openai/v1")

    print("=" * 50)
    print("LangChain + LLMSim Example")
    print("=" * 50)
    print(f"\nConnecting to: {base_url}")
    print()

    # Create ChatOpenAI client pointing to llmsim
    # Note: api_key is required by LangChain but llmsim doesn't validate it
    llm = ChatOpenAI(
        base_url=base_url,
        api_key="not-needed",  # llmsim doesn't require auth
        model="gpt-5",
        temperature=0.7,
        max_tokens=100,
    )

    # Example 1: Simple completion
    print("1. Simple Completion")
    print("-" * 30)
    messages = [
        SystemMessage(content="You are a helpful assistant."),
        HumanMessage(content="What is the capital of France?"),
    ]

    try:
        response = llm.invoke(messages)
        print(f"Response: {response.content}")
        print(f"Model: {response.response_metadata.get('model_name', 'unknown')}")
        if usage := response.response_metadata.get("token_usage"):
            print(f"Tokens: {usage}")
    except Exception as e:
        print(f"Error: {e}")
        print("\nMake sure the llmsim server is running:")
        print("  llmsim serve --port 8080")
        sys.exit(1)
    print()

    # Example 2: Streaming
    print("2. Streaming Response")
    print("-" * 30)
    print("Response: ", end="", flush=True)

    for chunk in llm.stream(messages):
        print(chunk.content, end="", flush=True)
    print("\n")

    # Example 3: Different models
    print("3. Different Models")
    print("-" * 30)

    models = ["gpt-5-mini", "claude-opus-4.5", "o3-mini"]
    for model in models:
        llm_model = ChatOpenAI(
            base_url=base_url,
            api_key="not-needed",
            model=model,
            max_tokens=50,
        )
        response = llm_model.invoke([HumanMessage(content="Hello!")])
        print(f"{model}: {response.content[:60]}...")
    print()

    # Example 4: Batch processing
    print("4. Batch Processing")
    print("-" * 30)
    batch_messages = [
        [HumanMessage(content="Say 'one'")],
        [HumanMessage(content="Say 'two'")],
        [HumanMessage(content="Say 'three'")],
    ]

    responses = llm.batch(batch_messages)
    for i, resp in enumerate(responses, 1):
        print(f"  Batch {i}: {resp.content[:40]}...")
    print()

    print("=" * 50)
    print("Examples complete!")
    print("=" * 50)


if __name__ == "__main__":
    main()