Goblin
News
AI news by
promptgoblins.ai
|
News
About
News
About
Filtered by:
llama.cpp
Clear
Titles
Summaries
6
Metal Capability Shim Unlocks Near-Bare-Metal LLM Inference in Apple Silicon macOS VMs, Up to 16x Faster
Research
1
Aug 11
6
Metal Capability Shim Unlocks Near-Bare-Metal LLM Inference in Apple Silicon macOS VMs, Up to 16x Faster
Research
· 1 src · Aug 11
Discuss
6
llama.cpp PR Adds Multi-Token Prediction for Qwen3-Next, Boosting Inference Speed Up to 74%
Open Source
1
Aug 4
6
llama.cpp PR Adds Multi-Token Prediction for Qwen3-Next, Boosting Inference Speed Up to 74%
Open Source
· 1 src · Aug 4
Discuss
6
llama.cpp Adds MTP and DSpark Speculative Decoding for DeepSeek V4 Flash, Delivering ~50% Inference Speedup
Products
1
Aug 2
6
llama.cpp Adds MTP and DSpark Speculative Decoding for DeepSeek V4 Flash, Delivering ~50% Inference Speedup
Products
· 1 src · Aug 2
Discuss
6
DSpark Speculative Decoding Added to llama.cpp, Improving Draft Acceptance via Markov Head
Open Source
1
Jul 29
6
DSpark Speculative Decoding Added to llama.cpp, Improving Draft Acceptance via Markov Head
Open Source
· 1 src · Jul 29
Discuss
Filters
Signal
Title
Category
Sources
Posted
Discuss
