Orthrus
PulseAugur coverage of Orthrus — every cluster mentioning Orthrus across labs, papers, and developer communities, ranked by signal.
-
New speculative decoding methods boost LLM inference speed · 7 sources tracked
Researchers are advancing speculative decoding techniques for large language models to improve inference speed. Two new arXiv papers, ECHO and LoopSpec, introduce hierarchical and pipelined approaches, respectively, to …
-
Orthrus to release Qwen 3.5/3.6 and Gemma 4 models with open-source code
Orthrus, a project focused on training large language models, is preparing to release support for Qwen 3.5, Qwen 3.6, and Gemma 4 models. Alongside the model checkpoints, Orthrus will also open-source its complete train…
-
AI tools advance code generation, UI design, and LLM inference
Several recent articles explore advancements and practical applications in AI for code generation and development workflows. One piece introduces Orthrus, a method for parallel token generation in LLMs that maintains ou…