Qwen 3.8-Flash-Next open-release tomorrow: 125B total, 6B active MoE model

Qwen 3.8-Flash-Next releasing tomorrow (125B a6B)

Qwen teased Qwen3.8-Flash-Next on ModelScope, a multimodal MoE model built on the next-gen Qwen4 architecture with 125B total and ~6B active parameters. The early release is meant to preview Qwen4's design for the community. It drops 2026-08-26 15:00 UTC, with an FP8 variant alongside. The post doesn't disclose benchmarks, inference speed, or specific multimodal capabilities—I'll hold judgment until the model card lands.

Why it matters: Qwen is previewing the Qwen4 architecture with a 125B-total / 6B-active MoE design — real new information with high attention in the Chinese open-source community. The deduction is because it's not open-sourced until tomorrow, and no benchmarks or inference speed data are avai...