AIThe Decoderabout 2 hours ago
Alibaba releases Qwen3.8-Flash-Next, targeting "ultimate cost efficiency"
1 min read
Alibaba's Qwen team previewed the Qwen4 architecture with Qwen3.8-Flash-Next, a mixture-of-experts model activating 6 of 125 billion parameters per token. At one-ninth the training cost, it beats DeepSeek-V4-Flash and Claude Opus 4.6 on coding and office benchmarks, intensifying pricing pressure on OpenAI and Anthropic.
Level
Hype check
Tap to vote and see what everyone thinks.
Summary by ByteBrief
