media Hugging Face Forums · 6d ago · open_models

Janas-LLM seeks testers for CPU-only MoE inference engine

from English

Janas-LLM is a new C-based, GPL-3.0 inference engine designed to run Mixture-of-Experts models larger than RAM on ordinary Linux machines without GPUs by streaming experts from the SSD.

The developer is requesting community testing to tune performance across diverse hardware configurations, specifically targeting AMD CPUs, Intel processors without AVX-VNNI, systems with 16 GB or less of memory, and SATA SSDs. The provided script automates building the engine, downloading Qwen models, verifying bit-identical arithmetic results, measuring speed on an idle machine, and reporting findings via GitHub issues.

This effort aims to improve the engine's self-tuning capabilities for threads, GPU usage, and expert loading strategies across a wider variety of hardware than the developer's single test machine.

Importance 1/3 Trust 1/3 Hugging Face Forums Inference efficiency Local inference

Read original

media Hugging Face Forums · just now · 1 view Live

AiiStream Q3.6 enables running Qwen3.6-35B-A3B on 8-16 GB Macs

The author has released AiiStream Q3.6, an open-source technique that allows the Qwen3.6-35B-A3B model to run on Apple Silicon devices with only 8 to 16 GB of RAM. By keeping routed experts on SSD and using early loading predictions, the method reduces MLX memory usage to approximately 1.7 GB while maintaining byte-identical outputs to the standard mlx_lm implementation.

github llama.cpp · 22h ago · 1 view

llama.cpp b11307 preserves batch order for speculative decoding

The llama.cpp project released build b11307, which fixes layer input ordering to preserve the original batch sequence during speculative decoding. This change ensures that token order is correctly maintained for unmasked NextN embeddings and compatible with tensor splits.

github llama.cpp · 1d ago · 1 view

llama.cpp b11301 fixes integer overflow in ggml/gguf

The llama.cpp project has released build b11301, which includes a fix for an integer overflow issue within the ggml and gguf components. This update addresses validation logic to prevent potential errors when handling zero-element tensors.

github llama.cpp · 1d ago · 3 views

llama.cpp b11302 fixes ThreadSanitizer data race in sparse indexer

The llama.cpp project released build b11302, which addresses a critical concurrency issue identified by ThreadSanitizer in the CI pipeline. The update resolves a data race within the sparse attention mechanism where multiple CPU threads were incorrectly writing to the same memory elements.

github llama.cpp · 1d ago · 2 views

llama.cpp b11298 adds dflash support to convert and feature extraction

The llama.cpp project has released build b11298, which introduces support for the dflash format. This update enables both conversion of models to this format and subsequent feature extraction capabilities.