Lab 2 — Your First API Call¶

⏱️ ~20 min

Call an LLM from code and watch temperature change the output. Open in Colab, or run locally with Python 3.11+.

In [ ]:

# Setup
import os, getpass
PROVIDER = "openai"  # or "anthropic"
if PROVIDER == "openai":
    !pip install -q openai
    if not os.environ.get("OPENAI_API_KEY"):
        os.environ["OPENAI_API_KEY"] = getpass.getpass("OpenAI API key: ")
else:
    !pip install -q anthropic
    if not os.environ.get("ANTHROPIC_API_KEY"):
        os.environ["ANTHROPIC_API_KEY"] = getpass.getpass("Anthropic API key: ")
print(f"Provider set to: {PROVIDER}")

# Setup import os, getpass PROVIDER = "openai" # or "anthropic" if PROVIDER == "openai": !pip install -q openai if not os.environ.get("OPENAI_API_KEY"): os.environ["OPENAI_API_KEY"] = getpass.getpass("OpenAI API key: ") else: !pip install -q anthropic if not os.environ.get("ANTHROPIC_API_KEY"): os.environ["ANTHROPIC_API_KEY"] = getpass.getpass("Anthropic API key: ") print(f"Provider set to: {PROVIDER}")

In [ ]:

def chat(messages, temperature=0.7, model=None):
    if PROVIDER == "openai":
        from openai import OpenAI
        client = OpenAI()
        resp = client.chat.completions.create(
            model=model or "gpt-4o-mini", messages=messages, temperature=temperature)
        return resp.choices[0].message.content
    else:
        from anthropic import Anthropic
        client = Anthropic()
        system = "".join(m["content"] for m in messages if m["role"] == "system")
        turns = [m for m in messages if m["role"] != "system"]
        resp = client.messages.create(
            model=model or "claude-3-5-haiku-20241022", max_tokens=1024,
            temperature=temperature, system=system or None, messages=turns)
        return resp.content[0].text

def chat(messages, temperature=0.7, model=None): if PROVIDER == "openai": from openai import OpenAI client = OpenAI() resp = client.chat.completions.create( model=model or "gpt-4o-mini", messages=messages, temperature=temperature) return resp.choices[0].message.content else: from anthropic import Anthropic client = Anthropic() system = "".join(m["content"] for m in messages if m["role"] == "system") turns = [m for m in messages if m["role"] != "system"] resp = client.messages.create( model=model or "claude-3-5-haiku-20241022", max_tokens=1024, temperature=temperature, system=system or None, messages=turns) return resp.content[0].text

A first call¶

In [ ]:

print(chat([{"role": "user", "content": "Give me a one-line tip for code reviews."}]))

print(chat([{"role": "user", "content": "Give me a one-line tip for code reviews."}]))

Temperature: 0 is repeatable, high is creative¶

Run each a few times and compare the variation.

In [ ]:

prompt = [{"role": "user", "content": "Invent a name for a coffee shop for programmers."}]
print("temp=0.0 :", chat(prompt, temperature=0.0))
print("temp=0.0 :", chat(prompt, temperature=0.0))
print("temp=1.0 :", chat(prompt, temperature=1.0))
print("temp=1.0 :", chat(prompt, temperature=1.0))

prompt = [{"role": "user", "content": "Invent a name for a coffee shop for programmers."}] print("temp=0.0 :", chat(prompt, temperature=0.0)) print("temp=0.0 :", chat(prompt, temperature=0.0)) print("temp=1.0 :", chat(prompt, temperature=1.0)) print("temp=1.0 :", chat(prompt, temperature=1.0))

Try it: change PROVIDER, swap the model, push temperature to 1.5. Notice temp=0 barely changes between runs; temp=1 varies a lot.

✅ Did it work?¶

• You saw a one-line code-review tip printed from a live model call.
• The two temp=0.0 coffee-shop names were identical (or nearly so), while the two temp=1.0 names differed — temperature visibly drove the variation.