// head_to_head
DeepSeek V4 Flash 0731 vs Qwen3.5-9B
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.
Qwen3.5-9B is the cheaper of the two; neither can be ranked on quality here.
Qwen3.5-9B 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.
deepseek
DeepSeek V4 Flash 0731
- Blended / 1M
- $0.190
- Context
- 1.3M
- Released
- Jul 31, 2026
- Overall score
- 74.2
qwen
Qwen3.5-9B
- Blended / 1M
- $0.112
- Context
- 262K
- Released
- Mar 10, 2026
- Overall score
- Not evaluated
Specs and pricing
| Metric | DeepSeek V4 Flash 0731 | Qwen3.5-9B |
|---|---|---|
| LiveBench overall Mean of the seven LiveBench category scores, 0–100. Higher is better. | 74.2 | — |
| Cost per point Measured benchmark spend divided by overall score — dollars per point of capability. | $0.0356 | — |
| Blended price / 1M 3:1 input:output mix, the usual shape of production traffic. | $0.190 | $0.112win |
| Input price / 1M | $0.040win | $0.100 |
| Output price / 1M | $0.640 | $0.150win |
| Cached input / 1M Price of an input token served from the prompt cache, where the provider publishes one. | $0.016 | — |
| Context window | 1.3Mwin | 262K |
| Max output tokens | 944Kwin | 33K |
Benchmarks by category
An overall score averages away the thing you probably care about. These are the seven LiveBench categories scored separately — DeepSeek V4 Flash 0731 on top, Qwen3.5-9B 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 | DeepSeek V4 Flash 0731 | Qwen3.5-9B |
|---|---|---|
| Support chatbot 1.2K in / 400 out × 200K requests | $59.07/mo | $36.00/mo |
| RAG assistant 8K in / 600 out × 100K requests | $60.80/mo | $89.00/mo |
| Coding agent 40K in / 4K out × 20K requests | $69.76/mo | $92.00/mo |
| Document extraction 20K in / 1.5K out × 50K requests | $86.80/mo | $111.25/mo |
| Bulk classification 500 in / 20 out × 5M requests | $152.00/mo | $265.00/mo |
Which should you pick?
You need to fit large documents in one call
DeepSeek V4 Flash 0731
Wider context window — 1.3M against 262K.
You are cost-constrained
Qwen3.5-9B
Cheaper on blended list price at $0.112 per million tokens.
DeepSeek V4 Flash 0731 vs Qwen3.5-9B FAQ
Which is better, DeepSeek V4 Flash 0731 or Qwen3.5-9B?
Qwen3.5-9B is the cheaper of the two; neither can be ranked on quality here. Qwen3.5-9B 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 DeepSeek V4 Flash 0731 cheaper than Qwen3.5-9B?
Qwen3.5-9B is cheaper. On a 3:1 input:output blend, DeepSeek V4 Flash 0731 lists at $0.190 per million tokens and Qwen3.5-9B at $0.112 — Qwen3.5-9B is 1.7× cheaper. Input and output are priced separately — DeepSeek V4 Flash 0731 charges $0.040 in and $0.640 out, Qwen3.5-9B charges $0.100 and $0.150 — so the model that looks cheaper flips depending on how output-heavy your workload is.
Does DeepSeek V4 Flash 0731 or Qwen3.5-9B have a bigger context window?
DeepSeek V4 Flash 0731 has the larger context window: 1.3M for DeepSeek V4 Flash 0731 against 262K for Qwen3.5-9B. 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 DeepSeek V4 Flash 0731 and Qwen3.5-9B support prompt caching?
DeepSeek V4 Flash 0731 publishes a cached-input rate of $0.016 per million tokens against a full input rate of $0.040. The catalogue lists no separate cached rate for Qwen3.5-9B, 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 2026-06-25, 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.