Hugging Face releases VLM speculative decoding drafter

Hugging Face released an experimental DSpark draft model for its LFM2.5-VL-3B vision-language model, adding 280M parameters (8.9%) for up to 3.13x faster decoding on device and 2.66x on an H100 without changing output quality, with day-one support for llama.cpp, MLX-VLM and SGLang.

• Qwengram-0.8B: n-gram memory transfer cuts perplexity 5.05%2026-09-25
• Transformers now runs llama.cpp quants2026-09-22
• OpenUI releases OUI-1 UI model2026-09-11
• AI Needs Guardrails: Leaders Call for Slowing Down2026-09-26
• AI Paradox Needs Guardrails, Not Panic2026-09-26