Press Space to continue
Press Space to continue
SITUATION EXPLAINED: Alibaba’s Qwen 3.8 Flash beats DeepSeek V4 Flash and Opus 4.6 Max on most tasks, at just $0.16/M input tokens. • It leads DeepSeek V4 Flash and Opus 4.6 Max on most tested tasks, at $0.16 input and $0.47 output per million tokens • 125B total parameters with just 6B active per token, plus a separate 51B N-gram embedding layer • That N-gram layer is the actual innovation. It stores common word groups as a phrase dictionary and can sit in ordinary system RAM rather than GPU…
Alibaba_Qwen@Alibaba_QwenAug 26
⚡Meet Qwen3.8-Flash, a multimodal MoE and an early preview of the Qwen4 architecture, now open-weight! The production version Qwen3.8-Flash will be available soon via QwenCloud API at just $ 0.16/1M input tokens and $ 0.47/1M output tokens. 125B parameters + 51B N-gram embeddings, with just 6B activated per token. Unmatched cost-efficiency. What's new: 🥳 - Next architecture: GDN + QSA hybrid attention, Gated Residual, N-gram Embedding & Muon optimizer, serving as a precursor to the architecture used in Qwen4. - Dramatically lower training and inference costs: trained at just 1/9 the cost of Qwen3.7-Plus, while outperforming it across the board with especially strong gains in coding and office tasks. - Strong performance: scoring 58.7 on DeepSWE 1.1, 62.5 on SWE-bench Pro, 73.9 on CoWorkBench, 84.5 on AndroidWorld, and 95.7 on MathVision (with CI). - 262K native context, extensible to 1M with YaRN. We’re also releasing the weights for Qwen3.8-Flash-Next, giving the community an early look at the new architecture we’re exploring for Qwen4.🚀 We can't wait to see what you build with Qwen3.8-Flash!👀👇 - Blog: https://t.co/M5hYypFLgJ - Technical Report: https://t.co/IF0gObIkQO - Hugging Face: https://t.co/6ow8QVAABt - ModelScope: https://t.co/tDOn2jNuFG
View