Inkling-2-Small
A 383B parameter vision and text model from thinkingmachines, with one MLX build at 4-bit.
- Parameters
- 383B
- Smallest download
- 335 GB
- Memory needed
- From 377 GB
- Downloads, 30 days
- 0
- First MLX build
- 13 Nov 2026
About
Inkling 2 Small is a general-purpose multimodal model that accepts text and image inputs and generates text outputs. It is intended for use in English and other languages, and across multiple coding languages. The model is designed to be used by developers building AI-powered applications, including agentic and tool-use systems, coding assistants, chatbots, and retrieval-augmented generation systems, and is suitable for general-purpose conversational use, instruction-following, and other natural language and visual tasks.
- Architecture
- Inkling2ForConditionalGeneration
Build
| ↗ | 4-bit | 335 GB | 377 GB | — | 0 |
|---|
Run Inkling-2-Small-oQ4e
Terminal · mlx-vlm
pip install -U mlx-vlm
python -m mlx_vlm.generate --model mlx-community/Inkling-2-Small-oQ4e --image photo.jpg --prompt "Describe this image"Similar models
Inkling-Small
Downloads, 30 days
857
Memory (est.)
101 GB
Download (est.)
From 88.4 GB
Devices
Runs on: Mac
Quantisation, bits per weight
apache-2.0Added 30 Jul 2026
Inkling
Downloads, 30 days
567
Memory (est.)
354 GB
Download (est.)
From 315 GB
Devices
Runs on: Mac
Quantisation, bits per weight
apache-2.0Added 16 Jul 2026
Inkling-NVFP4
Downloads, 30 days
108
Memory (est.)
652 GB
Download (est.)
581 GB
Devices
Runs on: Mac
Quantisation, bits per weight
apache-2.0Added 15 Jul 2026
Llama-4-Maverick-17B-128E
Downloads, 30 days
226
Memory (est.)
—
Download (est.)
—
Devices
Runs on: Mac
Quantisation, bits per weight
llama4Added 7 Apr 2025
History
- MLX 4-bit. Inkling-2-Small-oQ4e published on mlx-community.