GPT-4o vs GPT-Realtime-2.1.
model vs model · prices as of
GPT-4o takes 3 of 5 rounds against GPT-Realtime-2.1. It is also the cheaper of the two to run, at $16.63 per million tokens against $36.
GPT-4o
$16.63/M tokens
cheapest we can explain, at Jiekou · $2.00 per 1,000 pages
128k context · tool calling · reads images
GPT-Realtime-2.1
$36/M tokens
OpenAI list price, nothing below it is accounted for · $3.84 per 1,000 pages
128k context · tool calling · reads images
List price in (lower wins)
$5/M
$4/M
List price out (lower wins)
$15/M
$24/M
Cheapest we can explain (lower wins)
$16.63/M at Jiekou
none below list
Standard price (lower wins)
$30/M
$36/M
Cost per 1,000 pages (lower wins)
$2.00
$3.84
Context
128k tokens
128k tokens
Open weights
no
no
Tool calling
yes
yes
Structured output
yes
no
Reads images
yes
yes
Reasoning
no
yes
Released
13 May 2024
6 Jul 2026
Providers selling it
29
4
Price we measure against
the maker's own
the maker's own
Trains on your prompts (no is better)
not stated
no provider to check
Providers selling both · 1
OpenAI$30 / $36
GPT-4o then GPT-Realtime-2.1, per million tokens
Go further
Add a third or fourth model, or switch the weighting, in the tool.
What moved in the last seven days
- GPT-4o
Nous Portal$2 / $8 $2.50 / $1025% dearer ·
Frequently asked
- Which is cheaper, GPT-4o or GPT-Realtime-2.1?
- GPT-4o, at $16.63 per million tokens against $36 for GPT-Realtime-2.1, three parts input to one part output, read on 17 Sept 2026.
- Where is each one cheapest?
- GPT-4o: $16.63 at Jiekou. GPT-Realtime-2.1: nobody goes below the maker's own price with a reason we can point at, so $36 is the number to beat.
- Can I buy both from one provider?
- Yes. One provider sells both: OpenAI. The cheapest of them for the two together is OpenAI.
- Which holds more context?
- Both hold 128k tokens in one call.
Prices exclude tax and are per million tokens, read 17 Sept 2026 from each provider's own catalog. The standard price blends three parts input to one part output; a page is 600 tokens in and 60 out. Bars are scaled to the larger value in each row.