[Models](/models) / [Qwen 2.5](/models?family=Qwen%202.5#models-gallery) / FuseChat Qwen 2.5 7B Instruct ![Qwen logo](/model-logos/qwen.svg) # Run FuseChat Qwen 2.5 7B Instruct on iPhone, iPad & Mac FuseChat Qwen 2.5 7B Instruct runs 100% private on iPhone, iPad & Mac inside Private LLM — no internet connection required, no data sent to any server. 7.6BiPhoneiPadMac ## Download the quantized weights These are the exact GPTQ-Int4 weights Private LLM runs for FuseChat Qwen 2.5 7B Instruct, published on our Hugging Face org. They're standard weights you can load in any app that supports the format, not just Private LLM. - [iPhone & iPad — GPTQ-Int4 →](https://huggingface.co/numen-tech/FuseChat-Qwen-2.5-7B-Instruct-GPTQ-Int4) - [Mac — GPTQ-Int4 →](https://huggingface.co/numen-tech/FuseChat-Qwen-2.5-7B-Instruct-GPTQ-Int4) ## Specifications Parameters 7.6B Context window 33K tokens Quantization GPTQ-Int4 Family Qwen 2.5 ## What FuseChat Qwen 2.5 7B Instruct is good at general FuseChat 3.0 is a series of language models that combine the strengths of several larger, powerful models into smaller, more efficient ones. It is good at following instructions, general conversation, math, and coding tasks. The model shows notable performance gains, especially in instruction-following benchmarks. ## Which of your devices can run it ### iPhone iPhone 17 Pro MaxiPhone 17 ProiPhone AiriPhone 17iPhone 16 Pro / 16 Pro MaxiPhone 16 / 16 PlusiPhone 16eiPhone 15 Pro / 15 Pro Max ### iPad iPad Pro (M5, 16GB)iPad Pro (M5, 12GB)iPad Pro (M4, 16GB)iPad Pro (M4, 8GB)iPad Pro (M2, 16GB)iPad Pro (M2, 8GB)iPad Pro (M1, 16GB)iPad Pro (M1, 8GB)iPad Air (M4)iPad Air (M3)iPad Air (M2)iPad Air (M1)iPad mini (A17 Pro) ### Mac Mac (Apple Silicon, 192GB)MacBook Pro (M4 Max, 128GB)Mac Studio / Pro (Apple Silicon, 96GB)MacBook Pro (M4 Max, 64GB)MacBook Pro (M4 Max, 48GB)MacBook Pro (M4 Max, 36GB)Mac (Apple Silicon, 32GB)MacBook Air (M4, 24GB)MacBook Air (M-series, 16GB)Mac (Apple Silicon, 8GB) Browse every model that fits [iPhone](/models/best-for-iphone) or [Mac](/models/best-for-mac). ## How to run FuseChat Qwen 2.5 7B Instruct in Private LLM 1. Download Private LLM from the App Store. 2. Open the in-app model library and choose FuseChat Qwen 2.5 7B Instruct. 3. Download the model once, then chat fully offline. [![Download Private LLM on the App Store](/app-store/download-badge/en/download.svg)![Download Private LLM on the App Store](/app-store/download-badge/en/download-dark.svg)](/download)[Join our Discord](/discord) ## Variants & related models ![Qwen logo](/model-logos/qwen.svg) ### Dolphin 3.0 Qwen 2.5 0.5B 0.5B uncensored 33K contextiPhone & iPad · Mac [View details →](/models/dolphin3.0-qwen2.5-0.5b) ![Qwen logo](/model-logos/qwen.svg) ### Dolphin 3.0 Qwen 2.5 1.5B 1.5B uncensored 131K contextiPhone & iPad · Mac [View details →](/models/dolphin3.0-qwen2.5-1.5b) ![Qwen logo](/model-logos/qwen.svg) ### Dolphin 3.0 Qwen 2.5 3B 3.1B uncensored 33K contextiPhone & iPad · Mac [View details →](/models/dolphin3.0-qwen2.5-3b) ![Qwen logo](/model-logos/qwen.svg) ### EVA D Qwen2.5 1.5B v0.0 1.8B creative 131K contextiPhone & iPad · Mac [View details →](/models/eva-d-qwen2.5-1.5b-v0.0) ![Qwen logo](/model-logos/qwen.svg) ### EVA Qwen2.5 7B v0.1 7.6B creative 131K contextiPhone & iPad · Mac [View details →](/models/eva-qwen2.5-7b-v0.1) ![Qwen logo](/model-logos/qwen.svg) ### OpenHands LM 7B v0.1 7.6B coding 33K contextiPhone & iPad [View details →](/models/openhands-lm-7b-v0.1) ## Frequently asked questions - Can I run FuseChat Qwen 2.5 7B Instruct on iPhone? Yes. FuseChat Qwen 2.5 7B Instruct runs on iPhone models with enough memory, such as iPhone 17 Pro Max, iPhone 17 Pro, iPhone Air, iPhone 17, iPhone 16 Pro / 16 Pro Max, iPhone 16 / 16 Plus, iPhone 16e, iPhone 15 Pro / 15 Pro Max, fully on-device in Private LLM — no internet connection required. - Can I run FuseChat Qwen 2.5 7B Instruct on Mac? Yes. FuseChat Qwen 2.5 7B Instruct runs on Macs with enough unified memory, such as Mac (Apple Silicon, 192GB), MacBook Pro (M4 Max, 128GB), Mac Studio / Pro (Apple Silicon, 96GB), MacBook Pro (M4 Max, 64GB), MacBook Pro (M4 Max, 48GB), MacBook Pro (M4 Max, 36GB), Mac (Apple Silicon, 32GB), MacBook Air (M4, 24GB), MacBook Air (M-series, 16GB), Mac (Apple Silicon, 8GB), fully on-device in Private LLM. - Does FuseChat Qwen 2.5 7B Instruct work offline? Yes. Once downloaded in Private LLM, FuseChat Qwen 2.5 7B Instruct runs 100% on-device — no internet connection, and nothing is sent to any server. - Is FuseChat Qwen 2.5 7B Instruct free to use? Private LLM is a one-time purchase with no subscription and no per-message cost. The models themselves are open source — once downloaded, they run offline with nothing to pay per use. ## Why run FuseChat Qwen 2.5 7B Instruct in Private LLM Private LLM has run local AI on iPhone, iPad, and Mac since 2023, before Apple Intelligence existed. Inference happens on your device, so your FuseChat Qwen 2.5 7B Instruct conversations never reach a server. The part most apps gloss over is quantization, and that is exactly where on-device quality is won or lost. Most llama.cpp and MLX wrappers ship the same off-the-shelf 4-bit RTN weights. Private LLM ships GPTQ and OmniQuant quantization, tuned per model, and our 3-bit OmniQuant models match or beat those 4-bit RTN builds on the same Apple Silicon. Run the same model both ways and you feel it in the first reply. [See how our quantization works](/en/faq#Why-can-t-Private-LLM-load-models-directly-from-Hugging-Face). ### More from Numen [![Slop or Not app icon](/app-icons/slop-or-not.png) #### Slop or Not Catch AI-written text on iPhone and Mac. The check runs on-device, so nothing you paste gets uploaded. ](https://slopornot.ai)[![Clean Links app icon](/app-icons/clean-links.png) #### Clean Links Strip trackers and clutter from any link before you share it. The cleanup happens on your device. ](https://cleanlinks.app) Specifications and summary come from FuseChat Qwen 2.5 7B Instruct's [Hugging Face model card](https://huggingface.co/FuseAI/FuseChat-Qwen-2.5-7B-Instruct). Private LLM ships its own quantized models, built with GPTQ-Int4 quantization tuned per model, and isn't affiliated with the model's authors.