Qwen 3.8 2.4T-A95B API

Input / 1M tokens$1.51~~$1.71~~

Output / 1M tokens$4.53~~$5.14~~

LLMAlibaba12% below official984K contextPrompt caching

Loading playground…

About Qwen 3.8 2.4T-A95B

Qwen 3.8 2.4T-A95B by Alibaba is available on API Many from $1.51 per million input tokens and $4.53 per million output tokens, and you call it with the same API key you use for every other model.

It accepts OpenAI-style chat requests with streaming and tools, has a 984K context window and can return up to 131,072 tokens per response.

Prompt caching is supported: cached input tokens are billed at the lower cached rate shown in the price table.

You pay only for successful results: failed or rejected requests are refunded automatically, and every call is recorded in your console with its exact cost.

Related models

New−12%

Qwen 3.8 Maxqwen3.8-max

LLMAlibaba984K contextPrompt caching

$1.51 in · $4.53 outTry it

−12%

Qwen Turboqwen-turbo

LLMAlibaba

$0.264 in · $2.64 outTry it

−12%

Qwen 3.6 Flashqwen3.6-flash

LLMAlibaba

$0.151 in · $0.905 outTry it

−12%

Qwen 3.6 Plusqwen3.6-plus

LLMAlibaba

$0.252 in · $1.51 outTry it

−12%

Qwen 3.7 Flashqwen3.7-flash

LLMAlibaba984K contextPrompt caching

$0.025 in · $0.101 outTry it

−12%

Qwen 3.7 Maxqwen3.7-max

LLMAlibaba

$1.51 in · $4.53 outTry it

−12%

Qwen 3.7 Plusqwen3.7-plus

LLMAlibaba

$0.252 in · $1.01 outTry it

−12%

Qwen 3.8 27Bqwen3.8-27b

LLMAlibaba984K contextPrompt caching

$0.377 in · $1.51 outTry it

Frequently asked questions

How much does the Qwen 3.8 2.4T-A95B API cost?

Qwen 3.8 2.4T-A95B starts at $1.51 per million input tokens and $4.53 per million output tokens on API Many. The price table on this page lists every tier; you are charged only for successful requests.

How do I call Qwen 3.8 2.4T-A95B?

Create an API key in the console, then send a request to /v1/chat/completions with "model": "qwen3.8-2.4t-a95b". The API tab has copy-ready examples in cURL, Python and Node.js.

Is this the official Qwen 3.8 2.4T-A95B model?

Yes. Requests are forwarded to the model maker or an authorized channel with the same parameters, so the output matches the official API.

What happens if a request fails?

Failed, rejected or timed-out requests are not charged. Reserved balance is released automatically and the reason is shown in the call record.