Qwen API pricing
Qwen3.8 Max, the Alibaba (Qwen) flagship, costs $2 per million input tokens and $6 per million output tokens. Alibaba (Qwen)’s other current models, per million input / output tokens: Qwen3.8 27B $0.50 / $3, Qwen3.7 Plus $0.40 / $1.60 and Qwen3.8 Flash $0.15 / $0.47.
| Qwen3.8 Max | Qwen3.8 27B | Qwen3.7 Plus | Qwen3.8 Flash | |
|---|---|---|---|---|
| Input | $2.00 | $0.50 | $0.40 | $0.15 |
| Output | $6.00 | $3.00 | $1.60 | $0.47 |
US dollars per million tokens. List prices read 12 November 2026 from Alibaba (Qwen).
What the film says
- Paying list price for Claude?
- Claude Opus 5.5, Claude Fable 5.2, GPT-6 Astra, Gemini 4 Argon and Grok 4.7, +4 more models
- $1.60 instead of $4.00. Per million input tokens on Claude Opus 5.5.
- 9 of 9 models on sale pass the check against the same model at its maker. The results and the script are public: aigrow.gg/ai-api/genuine-models
- API credit$100 of API credit for $25. Start today.
Per token and per pass
The list price per million tokens, then how many of the benchmark’s 20 office tasks each model passed and what 1,000 passing answers cost, failed runs included.
| Model | Input | Output | Cached input | Context | Tasks passed | 1,000 passes |
|---|---|---|---|---|---|---|
| Qwen3.8 MaxFlagships | $2.00 | $6.00 | Not published | 1M | 17 of 20 | $16.32 |
| Qwen3.7 PlusWorkhorses | $0.40 | $1.60 | Not published | 1M | 17 of 20 | $3.58 |
| Qwen3.8 FlashSmall and fast | $0.15 | $0.47 | Not published | 1M | 19 of 20 | $1.09 |
| Qwen3.8 27BSmall and fast | $0.50 | $3.00 | Not published | 1M | 16 of 20 | $7.30 |
To see what your own workload costs on every model, use the LLM API cost calculator.
Alibaba (Qwen) prices the 4 current models on our list from Qwen3.8 Flash at $0.15 in and $0.47 out per million tokens to Qwen3.8 Max at $2 in and $6 out. On the AIGROW small-business benchmark Qwen3.8 Flash had the best record, 19 of 20 tasks passed, at $1.09 per thousand passes.
What three everyday jobs cost
1,000 runs of each job at list price. The token counts are ours; the prices are Alibaba (Qwen)’s.
| Model | 1,000 × support reply | 1,000 × document summary | 1,000 × agent task |
|---|---|---|---|
| Qwen3.8 Max | $8.40 | $56.00 | $178 |
| Qwen3.7 Plus | $1.84 | $11.60 | $36.80 |
| Qwen3.8 Flash | $0.64 | $4.22 | $13.41 |
| Qwen3.8 27B | $2.70 | $15.50 | $49.00 |
- Support reply
- A 3,000-token prompt (the ticket, the thread so far and a policy excerpt) and a 400-token answer.
- Document summary
- A 40-page document of about 25,000 tokens in, a 1,000-token summary out.
- Agent task
- Ten steps that read 80,000 tokens between them, 70,000 of them context repeated from step to step and served from cache, and write 3,000.
The small print
What Alibaba (Qwen)’s pricing pages add to the headline rates.
- Qwen3.8 Max
- International (Singapore) deployment, one tier up to 1M input tokens. Cache: implicit and explicit caching are discounted (explicit cache hits 10% of input, creation 125%), no single cached price recorded. Context taken from the top pricing tier (1M). Listed as a commercial model; Alibaba lists a separate open-source qwen3.8-2.4t-a95b.
- Qwen3.7 Plus
- Shown as 'List price (Limited-time 20% off)': the list price is recorded, the discounted rate would be $0.32 / $1.28. Input 256K to 1M: $1.20 / $4.80 list. Qwen3.7 Plus snapshot from 26 May 2026. International deployment. Cache discounts as for Qwen3.8 Max. Context taken from the top pricing tier (1M).
- Qwen3.8 Flash
- International deployment, one tier up to 1M input tokens. Cache discounts as for Qwen3.8 Max. Context taken from the top pricing tier (1M).
- Qwen3.8 27B
- Open-source model with a first-party hosted price (listed under 'Qwen (open source)'). International deployment, one tier up to 1M input tokens. Context taken from the top pricing tier (1M).
A price page for each Qwen model
Head to head
The comparisons with a Qwen model on one side, each with both prices and the three jobs costed on each.
Where the prices come from
- Alibaba (Qwen)
- alibabacloud.com/help/en/model-studio/model-pricing
Read on 12 November 2026. Each price is the standard rate for the shortest context band, with no batch or priority discount; long-context surcharges, batch rates and promotions are in each model’s small print. No price was taken from a reseller.
Weigh Qwen against every other maker
The price list sets the models of 11 makers side by side: input and output price per million tokens, and what three everyday jobs cost on each.
Prices checked 12 November 2026 · See them by release date · See the LLM leaderboard