Text Client

from django_cfg.modules.django_llm.client import LLMClient
 
client = LLMClient()
 
response = client.chat_completion(
    messages=[{"role": "user", "content": "Explain machine learning"}],
    model="openai/gpt-4o-mini"
)
print(response['content'])

from pathlib import Path
 
client = LLMClient(
    apikey_openrouter="sk-or-v1-...",
    apikey_openai="sk-proj-...",
    cache_dir=Path("cache/llm"),
    cache_ttl=3600,
    max_cache_size=1000
)

response = client.chat_completion(
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "What is AI?"}
    ],
    model="openai/gpt-4o-mini",
    temperature=0.7,
    max_tokens=500
)
print(response['content'])

for chunk in client.chat_completion_stream(
    messages=[{"role": "user", "content": "Tell me a story"}],
    model="openai/gpt-4o-mini"
):
    print(chunk, end='', flush=True)

functions = [
    {
        "name": "get_weather",
        "description": "Get current weather",
        "parameters": {
            "type": "object",
            "properties": {
                "location": {"type": "string"}
            }
        }
    }
]
 
response = client.chat_completion(
    messages=[{"role": "user", "content": "Weather in Paris?"}],
    model="openai/gpt-4o-mini",
    functions=functions,
    function_call="auto"
)

embedding = client.generate_embedding(
    text="Sample text for embedding",
    model="text-embedding-ada-002"
)
# Returns: List[float]

texts = ["Text 1", "Text 2", "Text 3"]
embeddings = client.generate_embeddings_batch(
    texts=texts,
    model="text-embedding-ada-002"
)
# Returns: List[List[float]]

def find_similar(query, documents):
    query_emb = client.generate_embedding(query)
 
    results = []
    for doc_id, doc_emb in documents.items():
        similarity = cosine_similarity(query_emb, doc_emb)
        results.append((doc_id, similarity))
 
    return sorted(results, key=lambda x: x[1], reverse=True)

data = {
    "users": [
        {"name": "John", "age": 30},
        {"name": "Jane", "age": 25}
    ]
}
 
toon_string = client.to_toon(data)
# Use in prompts for token savings

from django_cfg.modules.django_llm.tokenizer import Tokenizer
 
tokenizer = Tokenizer()
 
# Count tokens in text
count = tokenizer.count_tokens("Hello world", "gpt-4o-mini")
 
# Count tokens in messages
count = tokenizer.count_messages_tokens(messages, "gpt-4o-mini")

def manage_budget(messages, max_tokens=4000, model="gpt-4o-mini"):
    tokenizer = Tokenizer()
    current = tokenizer.count_messages_tokens(messages, model)
 
    while current > max_tokens and len(messages) > 1:
        messages.pop(1)  # Remove oldest, keep system
        current = tokenizer.count_messages_tokens(messages, model)
 
    return messages

from django_cfg.modules.django_llm.structured import JSONExtractor
 
extractor = JSONExtractor()
 
# Extract from response
json_data = extractor.extract_json_from_response(
    "Here's the data: {'name': 'John', 'age': 30}"
)

text = """
John Doe is a 30-year-old engineer in San Francisco.
"""
 
schema = {
    "type": "object",
    "properties": {
        "name": {"type": "string"},
        "age": {"type": "integer"},
        "location": {"type": "string"}
    }
}
 
result = extractor.extract_json(
    text=text,
    schema=schema,
    model="openai/gpt-4o-mini"
)
# {'name': 'John Doe', 'age': 30, 'location': 'San Francisco'}

texts = [
    "Alice, 25, designer",
    "Bob, 35, manager"
]
 
results = extractor.extract_json_batch(
    texts=texts,
    schema=schema,
    model="openai/gpt-4o-mini"
)

# Caching is automatic based on cache_dir and cache_ttl
client = LLMClient(
    cache_dir=Path("cache/llm"),
    cache_ttl=3600  # 1 hour
)
 
# Cache management
cache_info = client.get_cache_info()
print(f"Size: {cache_info['size']}")
print(f"Hit rate: {cache_info['hit_rate']:.2%}")

from django_cfg.modules.django_llm.registry import calculate_chat_cost
 
cost = calculate_chat_cost(
    model="openai/gpt-4o-mini",
    input_tokens=100,
    output_tokens=50,
    models_cache=models_cache
)
print(f"Cost: ${cost:.4f}")

info = client.get_client_info()
print(f"Cache dir: {info['cache_directory']}")
print(f"Models: {len(info['available_models'])}")

messages = [
    {"role": "system", "content": "You are a helpful assistant. Be concise."},
    {"role": "user", "content": "Explain quantum computing"}
]