GPT-5.4 Pro vs GPT-5.5 Pro
Prices checked
GPT-5.4 Pro and GPT-5.5 Pro split the 5 rounds this table counts: the maker's two prices, the cheapest price, the maker's blended price and context. The weighting you pick decides the rest.
OpenAIGPT-5.4 Pro
1.1M context · tool calling · reads images
$61.36/M tokensCheapest price, at Poe · $26.18 per 1,000 pages
OpenAIGPT-5.5 Pro
1.1M context · tool calling · reads images
$61.36/M tokensCheapest price, at Poe · $26.18 per 1,000 pages
Maker's price in (lower wins)
$30/M
$30/M
Maker's price out (lower wins)
$180/M
$180/M
Cheapest price (lower wins)
$61.36/M at Poe
$61.36/M at Poe
Maker's price, blended (lower wins)
$67.50/M
$67.50/M
Cost per 1,000 pages (lower wins)
$26.18
$26.18
Context
1.1M tokens
1.1M tokens
Open weights
no
no
Tool calling
yes
yes
Structured output
no
yes
Reads images
yes
yes
Reasoning
yes
yes
Released
5 Mar 2026
23 Apr 2026
Providers selling it
23
22
Price we measure against
the maker's own
the maker's own
Trains on your prompts (no is better)
not published
not published
Providers selling both · 18
Poe$61.36 / $61.36
Jiekou$64.13 / $67.50
Cloudflare AI Gateway$67.50 / $67.50
Eden AI$67.50 / $67.50
FastRouter$67.50 / $67.50
Impossibl$67.50 / $67.50
Kilo Gateway$67.50 / $67.50
LLM Gateway$67.50 / $67.50
GPT-5.4 Pro, then GPT-5.5 Pro, per million tokens; the 8 cheapest of 18.
Go further
Add a third or fourth model, or change the weighting, in the tool.
Frequently asked
- Which is cheaper, GPT-5.4 Pro or GPT-5.5 Pro?
- Neither: both run at $61.36 per million tokens, blended 3 in : 1 out, read on 31 Oct 2026.
- Where is each one cheapest?
- GPT-5.4 Pro: $61.36 at Poe. GPT-5.5 Pro: $61.36 at Poe.
- Can I buy both from one provider?
- Yes. 18 providers sell both: Poe, Jiekou, Cloudflare AI Gateway, Eden AI, FastRouter, Impossibl, and more. The cheapest of them for the two together is Poe.
- Which holds more context?
- Both hold 1.1M tokens in one call.
Prices exclude tax, read 31 Oct 2026 from each provider's own catalog. They are per M tokens, blended 3 in : 1 out, and a page is 600 tokens in and 60 out. Bars are scaled to the larger value in each row.
Related comparisons