Skip to content
Qwen logo

Run OpenHands LM 32B v0.1 on Mac

Autonomous software issue resolution based on Qwen 2.5 32B

OpenHands LM 32B v0.1 runs 100% private on Mac inside Private LLM — no internet connection required, no data sent to any server.

32.8BMaccoding

Model on Hugging Face

Specifications

Parameters
32.8B
Context window
33K tokens
License
MIT
Quantization
GPTQ-Int4
Family
Qwen 2.5 32B

What OpenHands LM 32B v0.1 is good at

OpenHands LM 32B v0.1 has 32 billion parameters, a 128K token context window for handling large codebases, and is built on Qwen Coder 2.5 Instruct 32B. It is fine-tuned using successful agent traces from the OpenHands software development agent to autonomously solve GitHub issues. The model focuses on software engineering tasks like issue resolution, achieving a 37.2% verified resolve rate on SWE-Bench Verified. It can run locally on hardware such as a single NVIDIA 3090 GPU.

Which of your devices can run it

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)

Browse every model that fits iPhone or Mac.

How to run OpenHands LM 32B v0.1 in Private LLM

  1. Download Private LLM from the App Store.
  2. Open the in-app model library and choose OpenHands LM 32B v0.1.
  3. Download the model once, then chat fully offline.
Download Private LLM on the App StoreJoin our Discord

Variants & related models

Qwen logo

EVA Qwen2.5 32B v0.2

32.8B

Roleplay and storywriting finetune of Qwen 2.5 32B with flexible tone control

131K contextMac
Qwen logo

Qwen 2.5 32B

32.8B

General-purpose instruction model excelling at coding and mathematics

33K contextMac
Qwen logo

Qwen 2.5 Coder 32B

32.8B

Coding specialist trained on 5.5T tokens, with 32k context for large files

33K contextMac
Mistral logo

CodeNinja 1.0 OpenChat 7B

7.2B

Code completion and refactoring specialist built on OpenChat 3.5

8K contextiPhone & iPad · Mac
Meta logo

Llama 3 WhiteRabbitNeo 8B v2.0

8B

Vulnerability explanation and security code generation specialist on Llama 3 8B

8K contextiPhone & iPad · Mac
Qwen logo

OpenHands LM 7B v0.1

7.6B

Solves GitHub issues and refactors code, built on Qwen 2.5

33K contextiPhone & iPad

Frequently asked questions

  • OpenHands LM 32B v0.1 is too large for current iPhones. It runs on Mac with enough unified memory inside Private LLM, fully offline.

  • Yes. OpenHands LM 32B v0.1 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), fully on-device in Private LLM.

  • Yes. Once downloaded in Private LLM, OpenHands LM 32B v0.1 runs 100% on-device — no internet connection, and nothing is sent to any server.

  • 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 OpenHands LM 32B v0.1 in Private LLM

Private LLM has run local AI on iPhone, iPad, and Mac since 2023, well before Apple Intelligence, LM Studio, Ollama, etc. existed. Inference happens on your device, so your OpenHands LM 32B v0.1 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.

Specifications and summary come from OpenHands LM 32B v0.1's Hugging Face model card, released under the MIT license. Private LLM ships its own quantized models, built with GPTQ-Int4 quantization tuned per model, and isn't affiliated with the model's authors.