Qwen3.8 2.4T A95B

2.4T OPEN TITAN

The largest open-weight MoE model in existence. 2.4 Trillion parameters with 95B activated per token and 1M context, available on HuggingFace and ModelScope.

Context

1,000,000 tokens

Architecture

2.4 Trillion Total (95B Activated)

Official API (1M)

$0.8 in / $2.4 out (Open)+ Weights Free to Self-Host

License

Qwen Community License

Published benchmarks · lab-reported

Vendor-published figures, not independently reproduced. Sources in Verification below.

SWE-BENCH

74.1%

MMLU-PRO

85.2%

Quickstart

curl -s https://modelregistry.tirup.in/api/cli?model=qwen-3-8-2-4t-a95b

OFFICIAL ANNOUNCEMENT WEIGHTS / REPOSITORY

BACK TO LEDGER

Sources

Qwen3.8 2.4T A95B open weights · Alibaba Cloud (Qwen)

Report an error in this record