Alibaba's omnimodal Qwen3.8 Flash-tier model: it reads text, images, audio and video and answers in text across a 1M-token window. it replaced Qwen3.8 Flash, retired 2026-10-06, at the same price.
pick it if
you need audio or video input at the Flash-tier price
you reuse long prompts: cache hits bill about a tenth of input
you want a 1M-token window well below Qwen3.8 Max's price
skip it if
you need audio or image output; it returns text only
Qwen4 Max is Alibaba's flagship Qwen4 model: a hosted 2.4T-parameter MoE that takes text, image and video, with a thinking mode. it succeeds Qwen3.8 Max at lower rates, and its weights are not published.
pick it if
you want Qwen4's flagship with image and video input
long agent sessions reuse a prefix; cache hits bill $0.075 per 1M, against $0.25 on Qwen3.8 Max
token-heavy agent runs; Alibaba reports 48 to 49% fewer tokens than Qwen3.8 Max on Terminal-Bench 4.0
skip it if
you need downloadable weights; Alibaba has announced open weights for smaller Qwen4 models, none for Max
you need audio input; Qwen3.8 Omni Flash takes audio, Max does not
faq
which is cheaper, Qwen3.8 Omni Flash or Qwen4 Max?
Qwen3.8 Omni Flash lists lower input and output prices. on a coding-agent session (2M tokens in, 60,000 out) Qwen3.8 Omni Flash costs $0.33 and Qwen4 Max $3.27.
can i switch between Qwen3.8 Omni Flash and Qwen4 Max mid-task?
yes. in an octomind session, /model octohub:qwen3.8-omni-flash and /model octohub:qwen4-max switch models without losing the conversation, so you can draft on one and review on the other.
try either model
switch between models inside the same octomind session.