Qwen3.8 Flash Next
Open-weight experimental preview of the Qwen4 architecture: hybrid-attention MoE (125B total, 6B active) with vision encoder for coding, agent tasks, and image and video understanding
BENCHMARKS
Not in the Epoch Capabilities Index. Epoch benchmarks what it can run, so this is a gap in coverage, not a poor result.
From the lab
Launch announcement · Aug 26, 2026Qwen3.8-Flash-Next: A New Architecture, Towards Ultimate Cost-EfficiencyRELEASE DATEAug 26, 2026
WEIGHTSOpen weights
CONTEXT WINDOW262.1K
MAX OUTPUT131.1K
REASONINGYes
TOOL CALLINGYes
INPUTStext, image, video
OUTPUTStext
Specifications from Models.dev. Availability and limits may differ by provider. Benchmark scores come from Epoch AI under CC BY 4.0.