How much does zuzu.codes cost?

The starter track is free — read all lessons and practice for free. Full access to every track (current and future) is $14.99/month. Cancel anytime.

How long does each track take?

Each track is designed as a 30-day challenge — one lesson per day, about 15 minutes each. Go at your own pace, but the structure is built around daily consistency.

What's the lesson format?

Each lesson is a student-teacher dialogue with code examples, followed by a hands-on code challenge in an in-browser editor. You read, you understand, then you write real code.

Do I need prior coding experience?

Our beginner track starts from absolute zero — no prior experience needed. Advanced tracks build on earlier ones, and the platform tells you exactly where to start.

How is zuzu.codes different from freeCodeCamp or Codecademy?

zuzu.codes uses a structured 30-day track format with dialogue-based teaching, an in-browser code editor, and gamification (XP, streaks, progress tracking). The format builds genuine understanding through daily practice.

Token cost awareness — Ai Foundations | zuzu.codes

Day 19 · ~11 min●

Every LLM call has a cost in tokens. Two prompts of different sizes burn different amounts. The result object has a .usage() method that surfaces the count.

python

result = Agent(model).run_sync("Tell me a fact about Python")
usage = result.usage()
print(f"requests: {usage.requests}")
print(f"input tokens: {usage.input_tokens}")
print(f"output tokens: {usage.output_tokens}")
print(f"total: {usage.total_tokens}")

Why care about token count when I'm on max-tier with monthly quota?

Three reasons:

Quota awareness — month resets. A 10K-token-per-call script burns the budget faster than a 1K-token script.
Latency — longer outputs take longer to generate. If you don't need 500 words, don't ask for 500 words.
Future-proofing — when you build production systems with paid AI, every token has a money cost. The habit of seeing the count makes that cost real.

Tokens — the unit of LLM cost

A token is roughly ~3-4 characters or ~0.75 words for English. Different models tokenize differently; the model decides what counts.

The `.usage()` shape

python

result = Agent(model).run_sync(prompt)
u = result.usage()
u.requests        # 1 (or more, for retries / tool calls)
u.input_tokens    # tokens in the prompt + system + history
u.output_tokens   # tokens in the response
u.total_tokens    # input + output

Roughly

English text: 1 token ≈ 0.75 word, so 100 words ≈ 130 tokens
Code: 1 token ≈ 2-3 characters, so a 1KB Python file ≈ 400 tokens
JSON: between text and code; punctuation eats tokens

Quick experiments

python

# short prompt, short response
short = Agent(model).run_sync("What is 2+2?")
print(short.usage().total_tokens)    # ~20

# long prompt, long response
long = Agent(model).run_sync("Explain Python's async/await syntax in 500 words")
print(long.usage().total_tokens)    # ~700

For lessons, keep an eye on the count — develop the habit before you're paying real money.

Cost levers

To reduce token cost:

Shorter prompt — drop unnecessary context
Shorter expected output — "in one sentence" vs "in 500 words"
Smaller model — the runtime can route to cheaper models when accuracy isn't critical (AI Advanced covers model routing)
Cache — repeat queries shouldn't re-call the model (AI Advanced)
Trim conversation history — long histories re-send everything every turn

Sign up to practice

Create a free account to get started. Paid plans unlock all tracks.

Day 19 · ~11 min●

Every LLM call has a cost in tokens. Two prompts of different sizes burn different amounts. The result object has a .usage() method that surfaces the count.

python

result = Agent(model).run_sync("Tell me a fact about Python")
usage = result.usage()
print(f"requests: {usage.requests}")
print(f"input tokens: {usage.input_tokens}")
print(f"output tokens: {usage.output_tokens}")
print(f"total: {usage.total_tokens}")

Why care about token count when I'm on max-tier with monthly quota?

Three reasons:

Quota awareness — month resets. A 10K-token-per-call script burns the budget faster than a 1K-token script.
Latency — longer outputs take longer to generate. If you don't need 500 words, don't ask for 500 words.
Future-proofing — when you build production systems with paid AI, every token has a money cost. The habit of seeing the count makes that cost real.

Tokens — the unit of LLM cost

A token is roughly ~3-4 characters or ~0.75 words for English. Different models tokenize differently; the model decides what counts.

The `.usage()` shape

python

result = Agent(model).run_sync(prompt)
u = result.usage()
u.requests        # 1 (or more, for retries / tool calls)
u.input_tokens    # tokens in the prompt + system + history
u.output_tokens   # tokens in the response
u.total_tokens    # input + output

Roughly

English text: 1 token ≈ 0.75 word, so 100 words ≈ 130 tokens
Code: 1 token ≈ 2-3 characters, so a 1KB Python file ≈ 400 tokens
JSON: between text and code; punctuation eats tokens

Quick experiments

python

# short prompt, short response
short = Agent(model).run_sync("What is 2+2?")
print(short.usage().total_tokens)    # ~20

# long prompt, long response
long = Agent(model).run_sync("Explain Python's async/await syntax in 500 words")
print(long.usage().total_tokens)    # ~700

For lessons, keep an eye on the count — develop the habit before you're paying real money.

Cost levers

To reduce token cost:

Shorter prompt — drop unnecessary context
Shorter expected output — "in one sentence" vs "in 500 words"
Smaller model — the runtime can route to cheaper models when accuracy isn't critical (AI Advanced covers model routing)
Cache — repeat queries shouldn't re-call the model (AI Advanced)
Trim conversation history — long histories re-send everything every turn

Tokens — the unit of LLM cost

The .usage() shape

Roughly

Quick experiments

Cost levers

Sign up to practice

Tokens — the unit of LLM cost

The .usage() shape

Roughly

Quick experiments

Cost levers

Sign up to practice

The `.usage()` shape

The `.usage()` shape