Qwen 3.8 API. 30 to 36% under Alibaba Cloud's international price.
Alibaba released Qwen3.8-Max on August 3, 2026. Call it and Qwen3.8-Flash with the OpenAI SDK on one APIsRouter key, no Alibaba Cloud account needed.
| Released | August 3, 2026 (Qwen3.8-Max) |
|---|---|
| Model IDs | qwen3.8-max, qwen3.8-flash |
| Context | Up to 1 million tokens |
Qwen 3.8 API prices, per million tokens
Alibaba Cloud's international price next to ours, for prompts up to 1 million tokens. Alibaba Cloud Model Studio international (Singapore) prices from alibabacloud.com/help/en/model-studio/model-pricing, September 2026. Alibaba Cloud shows Qwen 3.8 cache prices only in its console. On APIsRouter, cached input costs $0.18 per million on Qwen3.8-Max and $0.012 on Qwen3.8-Flash, and creating a cache costs $1.80 and $0.15.
| Model | Input: Alibaba Cloud / APIsRouter | Output: Alibaba Cloud / APIsRouter |
|---|---|---|
| Qwen3.8-Max (qwen3.8-max) | $2 / $1.40 | $6 / $4 |
| Qwen3.8-Flash (qwen3.8-flash) | $0.15 / $0.10 | $0.47 / $0.30 |
Max or Flash
- Qwen3.8-Max: Alibaba calls it its largest and most capable Qwen model yet. $1.40 / $4 per million input and output tokens on APIsRouter.
- Qwen3.8-Flash: The lower-priced Qwen 3.8 model. $0.10 / $0.30 per million input and output tokens on APIsRouter.
Call Qwen 3.8 in one line
- Get a key: Start free on the $2 trial credit, or top up from $10 below. A top-up key arrives by email.
- Point the OpenAI SDK at APIsRouter: Set base_url to https://api.apisrouter.com/v1 and use qwen3.8-max or qwen3.8-flash.
- Send requests: Every request shows up in your console with its tokens and cost.
from openai import OpenAI
client = OpenAI(
base_url="https://api.apisrouter.com/v1",
api_key="sk-your-apisrouter-key",
)
reply = client.chat.completions.create(
model="qwen3.8-max",
messages=[{"role": "user", "content": "Hello"}],
)
print(reply.choices[0].message.content)Two ways to pay
Both fill one balance, spent at the prices above.
FAQ
How much does the Qwen 3.8 API cost?
At Alibaba Cloud's international price, Qwen3.8-Max costs $2 / $6 per million input and output tokens and Qwen3.8-Flash $0.15 / $0.47. On APIsRouter, $1.40 / $4 and $0.10 / $0.30.
When was Qwen 3.8 released?
Alibaba Cloud previewed Qwen3.8-Max at WAIC on July 19, 2026 and released it on August 3, 2026.
Is Qwen 3.8 free?
On Alibaba Cloud's international (Singapore) site, each model comes with 1 million free tokens for 90 days from when you activate Model Studio or the model is released, whichever is later. On APIsRouter you can start on a $2 trial credit for both models: sign in and claim it from your dashboard, no card needed.
How long a prompt can Qwen 3.8 take?
Up to 1 million tokens, according to Alibaba Cloud.
Qwen3.8-Max or Qwen3.8-Flash?
Alibaba calls Qwen3.8-Max its most capable Qwen model; Qwen3.8-Flash costs less. On APIsRouter, Flash input costs $0.10 per million against $1.40 for Max, and the trial credit covers both, so you can compare them on your own prompts.
Does Qwen 3.8 work with the OpenAI SDK?
Yes. Set base_url (baseURL in Node) to https://api.apisrouter.com/v1 and use your APIsRouter key. On APIsRouter the Qwen models take OpenAI-format requests only, so use them from OpenAI-compatible tools rather than Claude Code.