#tHow Many Tokens?

← All models

GPT-5.1: token counter & pricing

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

Provider
OpenAI
API model ID
gpt-5.1
Context window
400,000 tokens
Input price
$1.25 per 1M tokens
Output price
$10.00 per 1M tokens
Tokenizer accuracy
exact (uses official tokenizer)
Pricing as of
2026-05-31

Open the counter to count tokens for GPT-5.1 in real time.

What is GPT-5.1?

GPT-5.1 is OpenAI's first incremental update to the original GPT-5, same tokenizer and context window, slightly improved instruction-following and coding behavior. Priced identically to GPT-5 itself at $1.25 input / $10 output per 1M tokens.

If you've already tuned a workload on GPT-5 and just want the quality bump without changing your cost model, GPT-5.1 is the drop-in.

How tokens are counted here

Uses OpenAI's o200k_base tokenizer (shared across the whole GPT-5 family + GPT-4o). Counts run in your browser via js-tiktoken, your prompt never leaves your device. Exact.

Pricing notes

$1.25 input / $10 output per 1M. Cached input drops to $0.125/M (10% standard) with OpenAI prompt caching.

Single-tier pricing, no long-context surcharge above 128K input the way GPT-5.4 / GPT-5.5 charge. That makes GPT-5.1 attractive for workloads that frequently push past 128K context.

When to use GPT-5.1

When not to use it:

Common questions

How does GPT-5.1 differ from GPT-5.2?

GPT-5.2 is priced higher ($1.75/$14) and adds further reasoning improvements. If you're paying GPT-5.2's premium, you should be measuring it winning on something specific, otherwise GPT-5.1 is the cheaper rational default in this tier.

Same context window as GPT-5?

Yes, 400K. Same as the whole GPT-5 family.

Will GPT-5.1 be deprecated?

OpenAI hasn't announced a sunset. The historical pattern: predecessor minor versions stay on the API for 12-24 months past the next flagship.

Compare GPT-5.1 to other models