llm-kb
← Back to research
Breakthroughfeatured

PrismML claims to run compressed 27B-parameter Qwen 3.6 model on iPhone 17 Pro

PrismML, a Khosla-backed startup, announced it has compressed Alibaba's open-source Qwen 3.6-27B model to run on an iPhone 17 Pro. This is claimed to be the largest-ever AI model deployed on a mobile device, as most on-device models have only a few billion active parameters. The achievement highlights advances in model compression for edge deployment.

76 engagement·1 source·Mon, Jul 13, 2026, 07:59 AM
PrismML claims to have shrunk Qwen 3.6, a 27-billion-parameter open-source LLM from Alibaba, to run on an iPhone 17 Pro. The startup says this is the largest AI model ever run on a phone, as typical mobile models have only a few billion active parameters. The compression technique is not detailed in the post. The claim was reported on Reddit on July 13, 2026, linking to an archived article. PrismML is backed by Khosla Ventures.

Entities

Alibaba(company)PrismML(company)Qwen 3.6-27B(model)Khosla Ventures(company)iPhone 17 Pro(tool)

Related

Model Release

Colibrì 744B-Parameter Model Runs on a Laptop

A new 744-billion-parameter model named Colibrì has been released, capable of running on a laptop. The model's name and parameter count suggest a focus on efficiency despite its large size.

0 engagement·1 source·rss
Sun, Jul 12, 2026, 08:50 AM
RSS
Tool Release

qMLX: Custom inference engine for Qwen 3.5 122B on Apple Silicon released

A developer released qMLX, a custom inference engine for Qwen 3.5 122B on Apple Silicon, extending MLX with hybrid attention support, SSD-backed KV cache, and RYS layer duplication. It enables efficient serving of the hybrid MoE model on consumer Mac hardware.

19 engagement·1 source·github
Thu, Jul 9, 2026, 09:47 PM
Model Release

Tencent Hy Team releases Hy3, a 295B-parameter MoE model

Tencent's Hy Team released Hy3, a 295-billion-parameter Mixture-of-Experts (MoE) language model with 21B active parameters and a 3.8B MTP layer. The model outperforms similar-sized models and rivals open-source models with 2-5x parameters. It is available on Hugging Face in full (598GB) and FP8 quantized (300GB) versions.

0 engagement·1 source·rss
Mon, Jul 6, 2026, 11:57 PM
RSS
Model Release

Cactus releases Needle, a 26M parameter function calling model

Cactus Compute released Needle, a 26 million parameter model specialized for function calling, claiming it outperforms much larger models. The model is open-source with a demo on Hugging Face and code on GitHub.

1.2k engagement·1 source·youtube
Sun, Jul 12, 2026, 01:00 PM
Community

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
Sun, Jul 12, 2026, 05:03 AM