Qwen 3.8 2.4T A95B TL1

TL1 Open weights

Frontier-scale coding, research, and long-horizon agents with up to 1M context

Pricing

Prices for this trust level. Prices are per 1M tokens.

Trust levels for this model
Trust Level Region Context Max outputInput / 1MCache / 1MOutput / 1M§203
TL1 SG — — 1.90 € 2.09 $ 0.22 € 0.24 $ 5.00 € 5.51 $ —

USD prices are converted from EUR at the ECB reference rate of 1 EUR = 1.1014 USD (2026-11-16). EUR prices are authoritative.

Prices as of 16.11.2026.

About this model

Qwen3.8 2.4T A95B is Alibaba's open-weight flagship reasoning model, built for demanding software engineering, professional research, and long-horizon agents. Its Mixture-of-Experts architecture has 2.4T parameters, activating 95B per token across 512 experts. A hybrid of Gated DeltaNet linear attention and full attention supports a 262K-token native context window, extendable to about 1M tokens, and up to 128K output tokens. The text-only model handles repository-scale coding, complex document analysis, multi-step research, and tool-driven automation. Thinking is always on; function calling and adjustable low, medium, or xhigh reasoning effort help agents plan, execute, and respond to feedback; preserved thinking can carry reasoning context across turns. The weights are released under the Qwen3.8-Max License, which permits commercial use subject to conditions.

API

Model ID qwen3.8-2.4t-a95b-tl1 OpenAI compatibleAnthropic compatible

Base URL https://api.aki.io/v1

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_AKI_API_KEY",
    base_url="https://api.aki.io/v1",
)

response = client.chat.completions.create(
    model="qwen3.8-2.4t-a95b-tl1",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_AKI_API_KEY",
  baseURL: "https://api.aki.io/v1",
});

const response = await client.chat.completions.create({
  model: "qwen3.8-2.4t-a95b-tl1",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);
curl https://api.aki.io/v1/chat/completions \
  -H "Authorization: Bearer $AKI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3.8-2.4t-a95b-tl1",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Frequently Asked Questions

How much does Qwen 3.8 2.4T A95B TL1 cost?
Input: 1.90 € per 1M tokens, Cache read: 0.22 € per 1M tokens, Output: 5.00 € per 1M tokens. Prices are exclusive of VAT.
What is the context window of Qwen 3.8 2.4T A95B TL1?
Qwen 3.8 2.4T A95B TL1 supports up to — tokens.
Which trust levels are available for Qwen 3.8 2.4T A95B TL1?
Qwen 3.8 2.4T A95B TL1 is available at TL1 (International Standard, SG).
Is Qwen 3.8 2.4T A95B TL1 open weights?
Yes. The model weights are available on Hugging Face under the linked license.
Is Qwen 3.8 2.4T A95B TL1 available with zero data retention?
Yes. Qwen 3.8 2.4T A95B TL1 is served at TL1 (International Standard) with zero data retention and no model training.
How do I use Qwen 3.8 2.4T A95B TL1 with the OpenAI SDK?
Use the model ID qwen3.8-2.4t-a95b-tl1 with the OpenAI-compatible endpoint at https://api.aki.io/v1.

Start Building with Qwen 3.8 2.4T A95B TL1