GPT-5.2 vs GPT-5.3 Codex
Prices checked
GPT-5.2 and GPT-5.3 Codex 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-5.2
400k context · tool calling · reads images
$4.38/M tokensCheapest price, at Poe · $1.72 per 1,000 pages
OpenAIGPT-5.3 Codex
400k context · tool calling · reads images
$4.38/M tokensCheapest price, at Poe · $1.72 per 1,000 pages
Maker's price in (lower wins)
$1.75/M
$1.75/M
Maker's price out (lower wins)
$14/M
$14/M
Cheapest price (lower wins)
$4.38/M at Poe
$4.38/M at Poe
Maker's price, blended (lower wins)
$4.81/M
$4.81/M
Cost per 1,000 pages (lower wins)
$1.72
$1.72
Context
400k tokens
400k tokens
Open weights
no
no
Tool calling
yes
yes
Structured output
yes
yes
Reads images
yes
yes
Reasoning
yes
yes
Released
11 Dec 2025
5 Feb 2026
Providers selling it
37
31
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 · 25
Poe$4.38 / $4.38
Jiekou$4.57 / $4.57
Abacus$4.81 / $4.81
AIHubMix$4.81 / $4.81
Azure AI Foundry$4.81 / $4.81
Azure Cognitive Services$4.81 / $4.81
Eden AI$4.81 / $4.81
FastRouter$4.81 / $4.81
GPT-5.2, then GPT-5.3 Codex, per million tokens; the 8 cheapest of 25.
Go further
Add a third or fourth model, or change the weighting, in the tool.
Frequently asked
- Which is cheaper, GPT-5.2 or GPT-5.3 Codex?
- Neither: both run at $4.38 per million tokens, blended 3 in : 1 out, read on 31 Oct 2026.
- Where is each one cheapest?
- GPT-5.2: $4.38 at Poe. GPT-5.3 Codex: $4.38 at Poe.
- Can I buy both from one provider?
- Yes. 25 providers sell both: Poe, Jiekou, Abacus, AIHubMix, Azure AI Foundry, Azure Cognitive Services, and more. The cheapest of them for the two together is Poe.
- Which holds more context?
- Both hold 400k 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