SenseTime has open-sourced SenseNova-Loop-9B-Preview, a multimodal model on a looped Transformer design it says can match models of about 27 billion parameters with only 9 billion. The card lists 32 decoder layers, a block of 24 run twice with shared weights, giving 56 layer passes per forward pass. Weights are on Hugging Face under Apache 2.0. https://pandaily.com/sensetime-sensenova-loop-9b-preview-looped-transformer-mhc-open-source-apache-2 #China #Tech #AI
Related
@zzt assuming you intended to allude to something similar, the most hilarious variant of this I've seen is this #iTunes ...
@zzt assuming you intended to allude to something similar, the most hilarious variant of this I've seen is this #iTunes comparison! :revblobfoxcrylaugh: #Slop #NoSlop #AI #NoAi #en...
私たちのフネにはジョリーが必要ですジョリーパスタ「リンゴのドルチェフェア」開催! 注文ごとに焼き上げるアップルパイなど3品 https://news.mynavi.jp/article/20261110-5277448/#Apple #LL...
私たちのフネにはジョリーが必要ですジョリーパスタ「リンゴのドルチェフェア」開催! 注文ごとに焼き上げるアップルパイなど3品 https://news.mynavi.jp/article/20261110-5277448/#Apple #LLM #news #bot
中野区ですか……タムズの次の仕事になりそうですauの基地局683局をAIと量子技術で再設定、混雑時の上り低速通信が約8割減 https://www.techno-edge.net/article/2026/11/10/5691.html#A...
中野区ですか……タムズの次の仕事になりそうですauの基地局683局をAIと量子技術で再設定、混雑時の上り低速通信が約8割減 https://www.techno-edge.net/article/2026/11/10/5691.html#Apple #LLM #news #bot