// head_to_head
GPT-5.2-Codex vs Qwen3.7 Max
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.
Effectively the same quality — Qwen3.7 Max is the cheaper way to get it.
The two are within 0.8 points overall, which is inside the range that effort settings alone move a LiveBench score, so treat them as quality-equivalent. Qwen3.7 Max lists 2.2× cheaper per blended million tokens. When quality ties, cost is the whole decision.
openai
GPT-5.2-Codex
- Blended / 1M
- $4.81
- Context
- 400K
- Released
- Jan 14, 2026
- Overall score
- 74.0
qwen
Qwen3.7 Max
- Blended / 1M
- $2.21
- Context
- 1M
- Released
- May 21, 2026
- Overall score
- 73.1
Specs and pricing
| Metric | GPT-5.2-Codex | Qwen3.7 Max |
|---|---|---|
| LiveBench overall Mean of the seven LiveBench category scores, 0–100. Higher is better. | 74.0 | 73.1 |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | $0.1001 | $0.0971 |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $4.81 | $2.21win |
| Input price / 1M | $1.75 | $1.48win |
| Output price / 1M | $14.00 | $4.42win |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $0.175win | $0.295 |
| Context window | 400K | 1Mwin |
| Max output tokens | 128K | 131K |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — GPT-5.2-Codex on top, Qwen3.7 Max 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-5.2-Codex | Qwen3.7 Max |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $1426.60/mo | $623.04/mo |
| RAG assistant 8K in / 600 out × 100K requests | $1610.00/mo | $973.50/mo |
| Coding agent 40K in / 4K out × 20K requests | $1638.00/mo | $873.20/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $2721.25/mo | $1747.88/mo |
| Bulk classification 500 in / 20 out × 5M requests | $4987.50/mo | $3540.00/mo |
Which should you pick?
The workload is coding or agentic work
GPT-5.2-Codex
Leads on agentic coding — 49.4 against 43.6.
You need to fit large documents in one call
Qwen3.7 Max
Wider context window — 1M against 400K.
GPT-5.2-Codex vs Qwen3.7 Max FAQ
Which is better, GPT-5.2-Codex or Qwen3.7 Max?
Effectively the same quality — Qwen3.7 Max is the cheaper way to get it. The two are within 0.8 points overall, which is inside the range that effort settings alone move a LiveBench score, so treat them as quality-equivalent. Qwen3.7 Max lists 2.2× cheaper per blended million tokens. When quality ties, cost is the whole decision.
Is GPT-5.2-Codex cheaper than Qwen3.7 Max?
Qwen3.7 Max is cheaper. On a 3:1 input:output blend, GPT-5.2-Codex lists at $4.81 per million tokens and Qwen3.7 Max at $2.21 — Qwen3.7 Max is 2.2× cheaper. Input and output are priced separately — GPT-5.2-Codex charges $1.75 in and $14.00 out, Qwen3.7 Max charges $1.48 and $4.42 — so the model that looks cheaper flips depending on how output-heavy your workload is.
GPT-5.2-Codex vs Qwen3.7 Max: which scores higher on benchmarks?
GPT-5.2-Codex scores 74.0 and Qwen3.7 Max scores 73.1 overall on LiveBench, the mean of its seven categories. That gap is inside the range that effort settings alone move a score, so treat them as equivalent on published quality. 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, GPT-5.2-Codex or Qwen3.7 Max?
They are close. GPT-5.2-Codex costs $0.1001 per point of overall capability and Qwen3.7 Max costs $0.0971, a difference small enough that workload shape will matter more than the rate.
Does GPT-5.2-Codex or Qwen3.7 Max have a bigger context window?
Qwen3.7 Max has the larger context window: 400K for GPT-5.2-Codex against 1M for Qwen3.7 Max. 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-5.2-Codex and Qwen3.7 Max support prompt caching?
Both publish a cached-input rate: $0.175 per million for GPT-5.2-Codex and $0.295 for Qwen3.7 Max, against full input rates of $1.75 and $1.48. 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.