Text Classification
MLX
Safetensors
English
Chinese
qwen3_5
computer-use
agents
decision-model
calibration
apple-silicon
local-llm
deskmind
8-bit precision
Instructions to use deskmind/brain-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use deskmind/brain-4b with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download deskmind/brain-4b --local-dir brain-4b
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Commit History
-
Model card: links to the Mac app, the training write-up and the hub repo
Verified signature84c59ca
-
main → G18b release (same files as g18b-q8)
Verified signature1409838 -
model card: G18b release (g18b-q8, router threshold 0.96, bench v25 39/39)
Verified signatureREADME only; weights and deskmind.json unchanged.f24afd5
-
model card: Jev comparison on the same suite (v23)
Verified signature54c2dd4 -
model card: pipeline tag and topics
Verified signaturebc803ba -
model card: g14-q8 release
Verified signaturee318d41 -
DeskMind Brain g14-q8 (G14, 8-bit MLX, prompt format 3)
Verified signature46e8a22 -
model card: g13-q8 release
Verified signatureee8f6ee -
DeskMind Brain g13-q8 (G13, 8-bit MLX, prompt format 3)
Verified signature674e86f -
model card, license, notice
Verified signatureeb08726 -
model card, license, notice
Verified signature1d96b13 -
model card, license, notice
Verified signatured17bfa6 -
DeskMind Brain v7b-q8 (G11b, 8-bit MLX)
Verified signature14c89f9 -
initial commit
Verified signatureb62c8ac