Alibaba (Qwen) › Qwen 3.5 9B › Pricing
$0.08 / $0.13per 1M tokens in / out
$0.21per task at thinking?
#108on OpenRouter by tokens · 60.1B / week
Price sheetper 1M tokens
| Input | $0.08 | Darkbloom |
| Output | $0.13 | |
| Context · max output | 262K · 33K | tokens |
Where to buy7 providers · per 1M tokens
| Provider | Input | Output | Context |
|---|---|---|---|
| Darkbloom | $0.08 | $0.13 | — |
| DeepInfra | $0.10 | $0.15 | — |
| Parasail | $0.10 | $0.25 | — |
| SiliconFlow | $0.10 | $0.15 | — |
| Venice | $0.10 | $0.15 | — |
| Featherless | $0.17 | $0.25 | — |
| Together | $0.17 | $0.25 | — |
In Alibaba (Qwen)'s line-upper 1M in / out
| Qwen 3.7 Flash 2026-07-15 | $0.03 / $0.13 | −41% |
| Qwen3.5 4B | $0.03 / $0.15 | −35% |
| Tongyi Intent Detect v3 | $0.058 / $0.14 | −14% |
| Qwen 3.5 9B | $0.08 / $0.13 | this model |
| Qwen Doc Turbo | $0.087 / $0.14 | +9% |
| Qwen 3.5 Flash | $0.065 / $0.26 | +23% |
| Qwen 3.5 OCR | $0.069 / $0.28 | +30% |
Cheaper for the same workvs Qwen 3.5 9B
| MiMo v2.6 Pro thinkingOpen | -37% | score ×4.1 | $0.13 |
| GPT-6.1 Sol low | -38% | score ×3.8 | $0.13 |
| Claude Haiku 5.5 xhigh | -42% | score ×3.7 | $0.12 |
| GPT-6 Luna max | -68% | score ×3.4 | $0.068 |
| MiMo v2.6 Flash thinkingOpen | -71% | score ×3.4 | $0.062 |
Pricing questions
How much does Qwen 3.5 9B cost?›
$0.08 per 1M input tokens and $0.13 per 1M output tokens. One Intelligence Index task costs about $0.21 at the default effort.
Is Qwen 3.5 9B free to use?›
There is no free API tier, but the weights are open: you can run it on your own hardware for the cost of compute.
Is Qwen 3.5 9B cheaper than Qwen3 VL 8B Instruct?›
Yes: $0.08 / $0.13 against $0.18 / $0.69, 70% less per token.
What is a cheaper alternative to Qwen 3.5 9B?›
MiMo v2.6 Pro at thinking effort costs -37% per task with a score ×4.1 on Humanity's Last Exam.