llm-kb
← Back to releases
Tool Release

Sparsify: run MoE models bigger than your RAM — Mixtral 26GB in 3.3GB RSS on a MacBook Air, byte-identical output

Sparsify, a new tool, enables running Mixtral 8x7B (26GB) on a MacBook Air with only 3.3GB RSS, producing byte-identical output. This is achieved by exploiting the sparsity of Mixture-of-Experts models, loading only the active experts into memory. The tool allows running models larger than available RAM without performance degradation.

0 engagement·1 source·Sun, Jul 12, 2026, 07:26 PM
Sparsify is a tool that leverages the inherent sparsity of Mixture-of-Experts (MoE) models to run them on devices with limited memory. In a demonstration, Mixtral 8x7B, which normally requires 26GB of RAM, was run on a MacBook Air using only 3.3GB of resident set size (RSS). The output was byte-identical to a full run, indicating no loss in quality. This is achieved by loading only the experts that are activated for a given input, rather than the entire model. The tool is particularly useful for running large MoE models on consumer hardware, such as laptops, without needing to quantize or prune the model. The post does not provide a download link or further technical details.

Entities

Sparsify(tool)Mixtral 8x7B(model)MacBook Air(tool)

Related

CommunitySun, Jul 12, 2026, 05:03 AM

Developer compresses GLM-5.2 MoE to run on single RTX 3090 via 79 experiments

A developer conducted 79 experiments to compress GLM-5.2, a 337 GB MoE model with 75 sparse layers and 256 routed experts, to fit on a 24 GB RTX 3090. The approach uses per-expert codecs, a batch pipeline over all MoE layers, and a patched llama.cpp runtime that loads codec-native expert binaries at inference time. The MIT-licensed repository documents the method and findings on expert similarity.

1 engagement·1 source·reddit
CommunitySun, Jul 12, 2026, 05:43 AM

User runs 100B+ MoE LLMs on low-end laptop using NVMe swap and Q3 quantization

A Reddit user with a low-spec laptop (i7-8750H, 20GB RAM, GTX 1050 4GB) reports successfully running 100B+ parameter MoE models by offloading parameters to a Samsung NVMe SSD via mmap, using Q3 quantization and quantized KV cache (Q4_0). They note that dense models are unusable on their hardware, but MoE models work with experts offloaded to CPU.

22 engagement·1 source·reddit
CommunityThu, Jul 9, 2026, 08:05 AM

User tests GLM 5.2 on consumer hardware, finds performance comparable to Claude and GPT

A user tested GLM 5.2 on a standard computer and was impressed by its capabilities and security, finding them similar to Claude or GPT. They began converting the model to int4 and exploring MTP usage to avoid out-of-memory errors.

1.1k engagement·1 source·hackernews
ProductSat, Jul 11, 2026, 10:45 PM

LLM hardware recipe database with filters and community usage tracking

A community-driven database that lists which LLM models run on which hardware, with performance details. Users can filter by hardware, submit new recipes, and mark which recipes they actively use to show popularity. It solves the problem of finding compatible model-hardware combinations for LLM deployment.

4 engagement·1 source·reddit
BenchmarkFri, Jul 10, 2026, 08:48 PM

Muse Spark 1.1 benchmarked against top models on Artificial Analysis

A user shared an Artificial Analysis comparison of Muse Spark 1.1 (xhigh) against models like Gemini 3.5 Flash, Claude Fable 5, and GPT-5.6 Sol, evaluating intelligence, performance, and cost per task. The benchmark provides practitioners with a data-driven view of where Muse Spark 1.1 stands relative to leading models.

38 engagement·1 source·reddit