35B Models
PulseAugur coverage of 35B Models — every cluster mentioning 35B Models across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
Lenovo Kaitian launches AI host with Zhaoxin CPU and Houmo NPU for local 35B models
Lenovo Kaitian has launched a compact AI host designed for local model execution. This device integrates the Zhaoxin KX-7000 CPU and the Houmo LQ50 NPU, offering 160 TOPS of performance. It is capable of running a 35-bi…
-
FreeToken environment enables 35B MoE models on 8GB GPUs
A new execution environment called FreeToken has been developed to enable large language models, specifically 35 billion parameter models, to run efficiently on GPUs with only 8GB of VRAM. This environment is optimized …
-
A.L.F.R.E.D. system distills large models into smaller ones for faster execution
A new system called A.L.F.R.E.D. (Adaptive Local-First Routing and Execution Distillation) has been developed to improve the efficiency of large language models. The approach involves distilling knowledge from larger mo…