#tHow Many Tokens?

← All models

o4-mini: token counter & pricing

OpenAI · exact (uses official tokenizer) · pricing as of 2026-05-31.

Provider
OpenAI
API model ID
o4-mini
Context window
200,000 tokens
Input price
$1.10 per 1M tokens
Output price
$4.40 per 1M tokens
Tokenizer accuracy
exact (uses official tokenizer)
Pricing as of
2026-05-31

Open the counter to count tokens for o4-mini in real time.

What is o4-mini?

o4-mini is OpenAI's newer-generation reasoning model, same approach as o3-mini (extended internal reasoning before responding) with broader reasoning improvements at the same price point: $1.10 input / $4.40 output per 1M tokens.

For new reasoning work where you don't need o3's full capacity, o4-mini is the rational default, same cost as o3-mini, better behavior.

How tokens are counted here

OpenAI's o200k_base tokenizer. Browser-side via js-tiktoken. Exact.

But: o4-mini generates "reasoning tokens", internal chain-of-thought tokens that count toward your output bill but don't appear in the final response. Typical 5-15× output overhead vs the visible reply. The calculator's per-call cost is visible-tokens only; budget several multiples of the output line for real-world spend.

Pricing notes

$1.10 input / $4.40 output per 1M. Cached input $0.275/M, 75% off standard, the best caching discount in the o-series.

200K context window.

For 1,000 input + 200 visible output (real billed output 1k-3k with reasoning):

When to use o4-mini

When not to use it:

Common questions

o4-mini vs o3-mini?

Same price, better behavior. For new work always pick o4-mini. o3-mini stays in the catalog for workloads pinned to its specific behavior profile.

o4-mini vs GPT-5.4?

GPT-5.4 ($2.50/$15) doesn't generate invisible reasoning tokens. For tasks where you don't actually need extended reasoning, GPT-5.4 is the cheaper effective choice. The break-even depends on how much reasoning your prompts actually require.

Is o4-mini still on the API after Feb 2026 ChatGPT retirement?

Yes, same situation as GPT-4o / GPT-4.1: retired from the ChatGPT UI, still callable via the API. No deprecation date published.

Compare o4-mini to other models