Total-Resort-3120

• ParoQuant: Pairwise Rotation Quantization for Efficient Reasoning LLM Inference

  Posted by Total-Resort-3120@reddit | LocalLLaMA | View on Reddit | 33 comments

• gemma-4-31B-it-DFlash has been released

  Posted by Total-Resort-3120@reddit | LocalLLaMA | View on Reddit | 33 comments

• ZAYA1-8B: Frontier intelligence density.

  Posted by Total-Resort-3120@reddit | LocalLLaMA | View on Reddit | 26 comments

• Ban phrases on llama.cpp with this script.

  Posted by Total-Resort-3120@reddit | LocalLLaMA | View on Reddit | 31 comments

• Want your LLM to use the internet? Here's an MCP server for that.

  Posted by Total-Resort-3120@reddit | LocalLLaMA | View on Reddit | 7 comments

• Let your LLM browse books locally so that it can write better stories.

  Posted by Total-Resort-3120@reddit | LocalLLaMA | View on Reddit | 1 comments

• DFlash: Block Diffusion for Flash Speculative Decoding.

  Posted by Total-Resort-3120@reddit | LocalLLaMA | View on Reddit | 127 comments

• DFlash: Block Diffusion for Flash Speculative Decoding.

  Posted by Total-Resort-3120@reddit | LocalLLaMA | View on Reddit | 0 comments