Released 9 days after Qwen3.8-Max · All Alibaba releases
Qwen3.8-2.4T-A95B is Alibaba's open-weight publication of its Max tier flagship, released in August 2026 and, at 2.4 trillion total parameters, the largest open-weight language model published to date. A sparse Mixture-of-Experts design activates roughly 95 billion parameters per token, paired with a hybrid attention stack that holds latency and compute well below a dense model of comparable size. This is the first time Alibaba has opened a Max tier model's weights, and the checkpoint is not identical to the managed service: the downloadable build is text only and always runs in thinking mode, while hosted Qwen3.8-Max adds vision input, a non-thinking mode and built-in tools. Native context is 262K tokens, extensible past 1 million, and it runs on standard inference stacks including vLLM and SGLang, which suits teams that want a frontier-scale open model they can host and adapt themselves.
Qwen3.8-2.4T-A95B runs on 3 different routes through the Opper gateway. Compare residency, ZDR, and training posture at a glance — full data-handling detail per route below.
Qwen3.8-2.4T-A95B runs on 2 independently monitored routes through the Opper gateway. A direct integration leaves you on one of them, the gateway serves whichever is healthy.
Measured over the last 30 days from each provider's official status feed via StatusGator. Refreshed hourly. 1 further route for this model is not on the board yet and is not counted above. See uptime for every provider Opper monitors.
Each route hosting Qwen3.8-2.4T-A95B has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.
Zero data retention is on by default on Pay-as-you-go — no action required. No training on customer data. US; unknown; DPA available.
Zero data retention is on by default on Pay-as-you-go — no action required. No training on customer data. GLOBAL; SCCs; DPA available.
Zero data retention is on by default on Pay-as-you-go — no action required. No training on customer data. EU; DPA available.
Independent benchmark scores — composite indices for reasoning, coding, and math, plus individual eval scores where available.
Call Qwen3.8-2.4T-A95B through the Opper gateway with one API key. Let your coding agent set it up, or call it directly — Opper is drop-in compatible with the OpenAI, Anthropic, and Google AI SDKs.
Copy this and paste it into a coding agent like Claude Code, Cursor or Codex and it'll wire up Opper for you.
import OpenAI from "openai";const client = new OpenAI({apiKey: process.env.OPPER_API_KEY,baseURL: "https://api.opper.ai/v3/compat",});const completion = await client.chat.completions.create({model: "deepinfra/Qwen/Qwen3.8-2.4T-A95B",messages: [{ role: "user", content: "Hello" }],});console.log(completion.choices[0].message.content);
Side-by-side on privacy, EU hosting, pricing, and benchmarks.
