// head_to_head
Claude Opus 4.5 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-5.2-Codex wins outright — it scores higher and costs less.
GPT-5.2-Codex leads by 1.4 points overall while listing 2.1× cheaper per blended million tokens. There is no trade-off to reason about here; the only reason to pick the other is a constraint this table does not show, like an existing contract, a region, or a provider you are already on. GPT-5.2-Codex also leads on measured cost per point of capability, at $0.1001 per point.
anthropic
Claude Opus 4.5
- Blended / 1M
- $10.00
- Context
- 200K
- Released
- Nov 24, 2025
- Overall score
- 72.6
openai
GPT-5.2-Codex
- Blended / 1M
- $4.81
- Context
- 400K
- Released
- Jan 14, 2026
- Overall score
- 74.0
Specs and pricing
| Metric | Claude Opus 4.5 | GPT-5.2-Codex |
|---|---|---|
| LiveBench overall Mean of the seven LiveBench category scores, 0–100. Higher is better. | 72.6 | 74.0win |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | $0.3211 | $0.1001win |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $10.00 | $4.81win |
| Input price / 1M | $5.00 | $1.75win |
| Output price / 1M | $25.00 | $14.00win |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $0.500 | $0.175win |
| Context window | 200K | 400Kwin |
| Max output tokens | 64K | 128Kwin |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Claude Opus 4.5 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 | Claude Opus 4.5 | GPT-5.2-Codex |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $2876.00/mo | $1426.60/mo |
| RAG assistant 8K in / 600 out × 100K requests | $3700.00/mo | $1610.00/mo |
| Coding agent 40K in / 4K out × 20K requests | $3480.00/mo | $1638.00/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $6650.00/mo | $2721.25/mo |
| Bulk classification 500 in / 20 out × 5M requests | $12,750/mo | $4987.50/mo |
Which should you pick?
You are running this at volume
GPT-5.2-Codex
Lowest measured cost per point of capability at $0.1001 per point — the gap compounds with every request.
Quality matters more than the bill
GPT-5.2-Codex
Highest overall LiveBench score of the two at 74.0.
The workload is coding or agentic work
GPT-5.2-Codex
Leads on agentic coding — 49.4 against 39.7.
You need to fit large documents in one call
GPT-5.2-Codex
Wider context window — 400K against 200K.
Claude Opus 4.5 vs GPT-5.2-Codex FAQ
Which is better, Claude Opus 4.5 or GPT-5.2-Codex?
GPT-5.2-Codex wins outright — it scores higher and costs less. GPT-5.2-Codex leads by 1.4 points overall while listing 2.1× cheaper per blended million tokens. There is no trade-off to reason about here; the only reason to pick the other is a constraint this table does not show, like an existing contract, a region, or a provider you are already on. GPT-5.2-Codex also leads on measured cost per point of capability, at $0.1001 per point.
Is Claude Opus 4.5 cheaper than GPT-5.2-Codex?
GPT-5.2-Codex is cheaper. On a 3:1 input:output blend, Claude Opus 4.5 lists at $10.00 per million tokens and GPT-5.2-Codex at $4.81 — GPT-5.2-Codex is 2.1× cheaper. Input and output are priced separately — Claude Opus 4.5 charges $5.00 in and $25.00 out, GPT-5.2-Codex charges $1.75 and $14.00 — so the model that looks cheaper flips depending on how output-heavy your workload is.
Claude Opus 4.5 vs GPT-5.2-Codex: which scores higher on benchmarks?
Claude Opus 4.5 scores 72.6 and GPT-5.2-Codex scores 74.0 overall on LiveBench, the mean of its seven categories. That is a 1.4-point lead for GPT-5.2-Codex. Category scores differ from the overall figure — a model can lead on reasoning and trail on coding, which the per-category table above breaks out.
Which gives better value for money, Claude Opus 4.5 or GPT-5.2-Codex?
GPT-5.2-Codex. Cost per point divides the measured dollars LiveBench spent running the benchmark by the score it earned, so it captures something token pricing misses: a reasoning model can emit many times more tokens than its per-token rate suggests. Claude Opus 4.5 works out at $0.3211 per point and GPT-5.2-Codex at $0.1001.
Does Claude Opus 4.5 or GPT-5.2-Codex have a bigger context window?
GPT-5.2-Codex has the larger context window: 200K for Claude Opus 4.5 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 Claude Opus 4.5 and GPT-5.2-Codex support prompt caching?
Both publish a cached-input rate: $0.500 per million for Claude Opus 4.5 and $0.175 for GPT-5.2-Codex, against full input rates of $5.00 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.