// head_to_head
GPT-4o (2024-11-20) vs GPT-5.2-Codex
Published LiveBench scores across all seven categories, live list pricing, context windows, and the measured cost of a point of capability — for both models, side by side.
GPT-4o (2024-11-20) and GPT-5.2-Codex are priced within ~10% of each other.
GPT-4o (2024-11-20) does not have a published LiveBench run, so this comparison covers price, context and declared capabilities only. A missing score means "not evaluated", not "worse" — the right way to separate these two is an eval on your own workload.
openai
GPT-4o (2024-11-20)
- Blended / 1M
- $4.38
- Context
- 128K
- Released
- Nov 20, 2024
- Overall score
- Not evaluated
openai
GPT-5.2-Codex
- Blended / 1M
- $4.81
- Context
- 400K
- Released
- Jan 14, 2026
- Overall score
- 74.0
Specs and pricing
| Metric | GPT-4o (2024-11-20) | GPT-5.2-Codex |
|---|---|---|
| LiveBench overall Mean of the seven LiveBench category scores, 0–100. Higher is better. | — | 74.0 |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | — | $0.1001 |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $4.38 | $4.81 |
| Input price / 1M | $2.50 | $1.75win |
| Output price / 1M | $10.00win | $14.00 |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $1.25 | $0.175win |
| Context window | 128K | 400Kwin |
| Max output tokens | 16K | 128Kwin |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — GPT-4o (2024-11-20) on top, GPT-5.2-Codex below, both out of 100.
What each one costs to run
Per-token prices are hard to feel. These are monthly list costs for both models across five workload shapes, using each provider's published cached-input rate where there is one.
| Workload | GPT-4o (2024-11-20) | GPT-5.2-Codex |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $1,310/mo | $1,427/mo |
| RAG assistant 8K in / 600 out × 100K requests | $2,100/mo | $1,610/mo |
| Coding agent 40K in / 4K out × 20K requests | $2,100/mo | $1,638/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $3,188/mo | $2,721/mo |
| Bulk classification 500 in / 20 out × 5M requests | $6,625/mo | $4,988/mo |
Which should you pick?
You need to fit large documents in one call
GPT-5.2-Codex
Wider context window — 400K against 128K.
GPT-4o (2024-11-20) vs GPT-5.2-Codex FAQ
Which is better, GPT-4o (2024-11-20) or GPT-5.2-Codex?
GPT-4o (2024-11-20) and GPT-5.2-Codex are priced within ~10% of each other. GPT-4o (2024-11-20) does not have a published LiveBench run, so this comparison covers price, context and declared capabilities only. A missing score means "not evaluated", not "worse" — the right way to separate these two is an eval on your own workload.
Is GPT-4o (2024-11-20) cheaper than GPT-5.2-Codex?
They cost about the same. Both land near $4.38 per million tokens on a 3:1 input:output blend, so price is unlikely to be the deciding factor between them.
Does GPT-4o (2024-11-20) or GPT-5.2-Codex have a bigger context window?
GPT-5.2-Codex has the larger context window: 128K for GPT-4o (2024-11-20) against 400K for GPT-5.2-Codex. Note that a window you can fill is not a window you should fill — retrieval quality usually degrades well before the limit, and you pay for every token you put in it.
Do GPT-4o (2024-11-20) and GPT-5.2-Codex support prompt caching?
Both publish a cached-input rate: $1.25 per million for GPT-4o (2024-11-20) and $0.175 for GPT-5.2-Codex, against full input rates of $2.50 and $1.75. On a workload with a long stable prefix — a system prompt, a tool schema, a retrieved corpus — that changes the economics more than the headline price does.
Related comparisons
How these numbers are produced
- Price — provider list price from OpenRouter, refreshed every 15 minutes. “Blended” is a 3:1 input:output mix.
- Scores — LiveBench release 2026-06-25, using their own category map. Each model shows its strongest published run. A blank means “not evaluated”, never “bad”.
- Cost per point — the measured dollars LiveBench spent on the run, divided by the score it earned.
- “Win” — awarded only past a threshold: one full point on a benchmark score, 10% on a price, 25% on a context window. Anything tighter reports as a tie, because effort settings alone move a LiveBench score by more than that.
Published benchmarks rank models on someone else's tasks. Before committing, see LLM & agent evaluation for building an eval on your own.