// head_to_head
Qwen3.5 Plus 2026-02-15 vs Qwen3.8 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.
Qwen3.5 Plus 2026-02-15 and Qwen3.8 27B are priced within ~10% of each other.
Qwen3.5 Plus 2026-02-15 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
Qwen3.5 Plus 2026-02-15
- Blended / 1M
- $0.585
- Context
- 1M
- Released
- Feb 16, 2026
- Overall score
- Not evaluated
qwen
Qwen3.8 27B
- Blended / 1M
- $0.581
- Context
- 1.0M
- Released
- Aug 14, 2026
- Overall score
- 58.4
Specs and pricing
| Metric | Qwen3.5 Plus 2026-02-15 | Qwen3.8 27B |
|---|---|---|
| LiveBench overall Weighted mean of the seven LiveBench category scores, 0–100. Higher is better. | — | 58.4 |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | — | $0.0084 |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $0.585 | $0.581 |
| Input price / 1M | $0.260 | $0.150win |
| Output price / 1M | $1.56win | $1.88 |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | — | $0.037 |
| Context window | 1M | 1.0M |
| Max output tokens | 66K | 236Kwin |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — Qwen3.5 Plus 2026-02-15 on top, Qwen3.8 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 | Qwen3.5 Plus 2026-02-15 | Qwen3.8 27B |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $187.20/mo | $177.90/mo |
| RAG assistant 8K in / 600 out × 100K requests | $301.60/mo | $187.50/mo |
| Coding agent 40K in / 4K out × 20K requests | $332.80/mo | $207.00/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $377.00/mo | $285.00/mo |
| Bulk classification 500 in / 20 out × 5M requests | $806.00/mo | $506.25/mo |
Qwen3.5 Plus 2026-02-15 vs Qwen3.8 27B FAQ
Which is better, Qwen3.5 Plus 2026-02-15 or Qwen3.8 27B?
Qwen3.5 Plus 2026-02-15 and Qwen3.8 27B are priced within ~10% of each other. Qwen3.5 Plus 2026-02-15 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 Qwen3.5 Plus 2026-02-15 cheaper than Qwen3.8 27B?
They cost about the same. Both land near $0.585 per million tokens on a 3:1 input:output blend, so price is unlikely to be the deciding factor between them.
Does Qwen3.5 Plus 2026-02-15 or Qwen3.8 27B have a bigger context window?
They are effectively the same — 1M for Qwen3.5 Plus 2026-02-15 and 1.0M for Qwen3.8 27B.
Do Qwen3.5 Plus 2026-02-15 and Qwen3.8 27B support prompt caching?
Qwen3.8 27B publishes a cached-input rate of $0.037 per million tokens against a full input rate of $0.150. The catalogue lists no separate cached rate for Qwen3.5 Plus 2026-02-15, which means the provider does not price it separately here — not that caching is unavailable.
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 2026-10-25, 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.