DeepSeek's MoE model with 552B total parameters and 8-16B active per token. 1M-token context, MIT license: free commercial use. Built for online serving, too large for local use on a single PC.
Back to resources
Models🀄
Qwen3.8-2.4T-A95B
Released on Nov 7, 2026
First Qwen-Max-class model in open weight: 2.4T total parameters with 95B active per token, a 512-expert MoE (10 routed + 1 shared), Gated DeltaNet + Gated Attention hybrid attention, 92 layers, native 262k-token context extendable to 1M. Separate FP8 variant.
Features
- MoE 2.4T/95B
- 512 experts (10+1)
- Gated DeltaNet
- 262k→1M context
- Open weight
Tags
Useful links
Discover Qwen3.8-2.4T-A95B
Visit the official site to download, learn more and start using Qwen3.8-2.4T-A95B.
Related resources — Models
Models⚡
DeepSeek V4.1 Flash
Released Sep 10, 2026DeepSeek V4.1 Flash
LLMDeepSeekMoEOpen Source
Models⚡
FLUX.1 [dev]
Released Aug 1, 2024FLUX.2 [klein] 4B/9B
The most advanced open image model: excellent prompt adherence, solid anatomy, full fine-tuning. The standard for serious work.
ImmaginiDiffusionFLUX
Models⚡
FLUX.1 [schnell]
Released Aug 1, 2024FLUX.2 [klein] 4B/9B
Very few steps, high quality, and strong text rendering. The distilled FLUX variant, built for speed.
ImmaginiDiffusionFLUX

