Qwen4 MaxvsStep 5 Preview
They share the same context, input modalities, output modalities, and reasoning, and differ in input price ($1.413 / 1M tokens vs $1.00 / 1M tokens), output price ($4.24 / 1M tokens vs $2.70 / 1M tokens), and input price (¥10.00 / 1M tokens vs ¥7.00 / 1M tokens).
Compare at your usage
Every model priced for the same request8,000 tokens in, 2,000 out
| Qwen4 Max on Alibaba Cloud Model Studio | Step 5 Preview on StepFun API | |
|---|---|---|
| Rate applied | Base rate | Base rate |
| Inputper 1M tokens | $1.413 | $1.00 |
| Cache readper 1M tokens | No cache price | $0.10 |
| Outputper 1M tokens | $4.24 | $2.70 |
| Cost per request | $0.01978 | $0.0134 |
| Per 1,000 requests | $19.78 | $13.40 |
- Qwen4 Max Alibaba Cloud Model Studio pricing · as of Nov 16, 2026
- Step 5 Preview StepFun: pricing details · as of Nov 7, 2026
Bold marks the cheapest for this request at each maker’s own list prices.
Cost per request as prompts grow
2K tokens out, list price on each maker’s API.
- Qwen4 Max · Alibaba Cloud Model Studio
- Step 5 Preview · StepFun API
Key differences
InputMaker’s official API · global · USD
$1.413 / 1M tokens Alibaba Cloud Model Studio pricing · as of Nov 16, 2026
$1.00 / 1M tokens StepFun: pricing details · as of Nov 7, 2026
OutputMaker’s official API · global · USD
$4.24 / 1M tokens Alibaba Cloud Model Studio pricing · as of Nov 16, 2026
$2.70 / 1M tokens StepFun: pricing details · as of Nov 7, 2026
InputMaker’s official API · china · CNY
¥10.00 / 1M tokens 阿里云百炼:模型价格in Chinese · as of Nov 16, 2026
¥7.00 / 1M tokens 阶跃星辰:定价明细in Chinese · as of Nov 9, 2026
OutputMaker’s official API · china · CNY
¥30.00 / 1M tokens 阿里云百炼:模型价格in Chinese · as of Nov 16, 2026
¥20.00 / 1M tokens 阶跃星辰:定价明细in Chinese · as of Nov 9, 2026
Max output
262,144 如何在 QwenPaw 中使用大模型 API-大模型服务平台百炼(Model Studio)-阿里云帮助中心in Chinese · as of Nov 16, 2026
64,000 Step 5 Preview · as of Nov 8, 2026
Full comparison
| Qwen4 Max Alibaba Qwen | Step 5 Preview StepFun | |
|---|---|---|
| Overview | ||
| Released | Nov 16, 2026 | Not stated |
| Open weights | — | |
| Specs | ||
| Context | 1M | 1M |
| Max output | 262.14K | 64K |
| Input modalities | Text, Image, Video | Text, Image, Video |
| Output modalities | Text | Text |
| Reasoning | ||
| Reasoning efforts | low, high, xhigh | low, medium, high |
| Reasoning can be off | Not stated | |
| Max thinking budget | 262.14K | Not stated |
| Tool use | ||
| Structured output | ||
| Pricing · per 1M tokens | ||
| Input price | $1.413 | $1.00 |
| Output price | $4.24 | $2.70 |
| Cache read price | — | $0.10 |
| Blended price (3 input : 1 output) | $2.11975 | $1.425 |
| Cited scores | ||
| LMArena Text | — | 1,448 |
| LMArena Text · Coding | — | 1,505 |
| LMArena Vision | — | 1,264 |
| LMArena Text · Chinese | — | 1,499 |
Missing facts stay blank; check each source before choosing a model. How we compare →
Model details and sources
- Qwen4 Max → Available
- Step 5 Preview → Preview
More comparisons with these models
All comparisons →Cited scores
- LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · LMArena leaderboard dataset (Hugging Face) · CC BY 4.0 · as of Nov 8, 2026