// head_to_head
Qwen2.5 VL 72B Instruct vs Qwen3.6 27B
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.
Qwen2.5 VL 72B Instruct and Qwen3.6 27B are priced within ~10% of each other.
Qwen2.5 VL 72B Instruct 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.
qwen
Qwen2.5 VL 72B Instruct
- Blended / 1M
- $0.850
- Context
- 128K
- Released
- Feb 1, 2025
- Overall score
- Not evaluated
qwen
Qwen3.6 27B
- Blended / 1M
- $0.915
- Context
- 262K
- Released
- Apr 27, 2026
- Overall score
- 64.0
Specs and pricing
| Metric | Qwen2.5 VL 72B Instruct | Qwen3.6 27B |
|---|---|---|
| LiveBench overall Mean of the seven LiveBench category scores, 0–100. Higher is better. | — | 64.0 |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | — | $0.1074 |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $0.850 | $0.915 |
| Input price / 1M | $0.800 | $0.320win |
| Output price / 1M | $1.00win | $2.70 |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $0.400 | $0.150win |
| Context window | 128K | 262Kwin |
| Max output tokens | 115K | 262Kwin |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Qwen2.5 VL 72B Instruct on top, Qwen3.6 27B 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 | Qwen2.5 VL 72B Instruct | Qwen3.6 27B |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $243.20/mo | $280.56/mo |
| RAG assistant 8K in / 600 out × 100K requests | $540.00/mo | $350.00/mo |
| Coding agent 40K in / 4K out × 20K requests | $496.00/mo | $376.80/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $855.00/mo | $514.00/mo |
| Bulk classification 500 in / 20 out × 5M requests | $1,900/mo | $985.00/mo |
Which should you pick?
You need to fit large documents in one call
Qwen3.6 27B
Wider context window — 262K against 128K.
Qwen2.5 VL 72B Instruct vs Qwen3.6 27B FAQ
Which is better, Qwen2.5 VL 72B Instruct or Qwen3.6 27B?
Qwen2.5 VL 72B Instruct and Qwen3.6 27B are priced within ~10% of each other. Qwen2.5 VL 72B Instruct 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 Qwen2.5 VL 72B Instruct cheaper than Qwen3.6 27B?
They cost about the same. Both land near $0.850 per million tokens on a 3:1 input:output blend, so price is unlikely to be the deciding factor between them.
Does Qwen2.5 VL 72B Instruct or Qwen3.6 27B have a bigger context window?
Qwen3.6 27B has the larger context window: 128K for Qwen2.5 VL 72B Instruct against 262K for Qwen3.6 27B. 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 Qwen2.5 VL 72B Instruct and Qwen3.6 27B support prompt caching?
Both publish a cached-input rate: $0.400 per million for Qwen2.5 VL 72B Instruct and $0.150 for Qwen3.6 27B, against full input rates of $0.800 and $0.320. 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.