llama.cpp PR adds DSpark speculative decoding
Pull request #25173 adds DSpark speculative decoding to llama.cpp; the author invites the community to benchmark and share pp/tg speedups. The PR links DeepSeek's DeepSpec collection, tying DSpark to DeepSeek-V4.
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- CPUs and the rise of neurosymbolic AI
- Mollick asks for plan on cyber threats from open-weight Mythos/Astra models
- Dwarkesh Patel discusses the implications of continual learning for AI
- NoimosAI launches Social Agent, an AI research agent for social trends
- Stanford researchers deploy 37,000 AI agents as a virtual biotech