o3 Mini High vs o4-mini.
model vs model · prices as of
o3 Mini High and o4-mini split the 5 rounds this table counts: the two list price rows, the cheapest price we can explain, the standard price and context. The weighting you pick decides the rest.
OpenAIo3 Mini High
$7/M tokens
cheapest we can explain, at Poe · $0.84 per 1,000 pages
200k context · tool calling
List price in (lower wins)
$1.10/M
$1.10/M
List price out (lower wins)
$4.40/M
$4.40/M
Cheapest we can explain (lower wins)
$7/M at Poe
$7/M at Poe
Standard price (lower wins)
$7.70/M
$7.70/M
Cost per 1,000 pages (lower wins)
$0.84
$0.84
Context
200k tokens
200k tokens
Open weights
not stated
no
Tool calling
yes
yes
Structured output
yes
yes
Reads images
no
yes
Reasoning
yes
yes
Released
12 Feb 2025
16 Apr 2025
Providers selling it
7
23
Price we measure against
the maker's own
the maker's own
Trains on your prompts (no is better)
not stated
not stated
Providers selling both · 7
Poe$7 / $7
FastRouter$7.70 / $7.70
Kilo Gateway$7.70 / $7.70
NanoGPT$7.70 / $7.70
Nous Portal$7.70 / $7.70- OpenAI$7.70 / $7.70
OpenRouter$7.70 / $7.70
o3 Mini High then o4-mini, 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
- o3 Mini HighNous Portal$0.88 / $3.52 $1.10 / $4.4025% dearer ·
- o4-miniNous Portal$0.88 / $3.52 $1.10 / $4.4025% dearer ·
- o4-mini
Jiekou$1.05 / $4.18 $1.10 / $4.405% dearer ·
Frequently asked
- Which is cheaper, o3 Mini High or o4-mini?
- Neither: both run at $7 per million tokens, three parts input to one part output, read on 16 Sept 2026.
- Where is each one cheapest?
- o3 Mini High: $7 at Poe. o4-mini: $7 at Poe.
- Can I buy both from one provider?
- Yes. 7 providers sell both: Poe, FastRouter, Kilo Gateway, NanoGPT, Nous Portal, OpenAI, and more. The cheapest of them for the two together is Poe.
- Which holds more context?
- Both hold 200k tokens in one call.
Prices exclude tax and are per million tokens, read 16 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.