mlxradar

MLX and Core AI models

Every model that runs on Apple silicon, with its download size and the memory it needs.

Popular Fits in 16 GB Core AI Vision models Both formats

2,359 models+8 this week34 in both formatsUpdated 23 minutes ago

Models

2,359 models
AllMLXCore AIBoth
TextVisionAudioImage
NVIDIA · 627M
parakeet-tdt-0.6b-v3
MLX Audio
Downloads, 30 days
2.7M
Memory (est.)
5 GB
Download (est.)
2.5 GB
Devices
Runs on: Mac
Quantisation not stated
cc-by-4.0Added 16 Aug 2025
NVIDIA · 618MTrending
parakeet-tdt-0.6b-v2
MLX Audio
Downloads, 30 days
847K
Memory (est.)
5 GB
Download (est.)
2.5 GB
Devices
Runs on: Mac
Quantisation not stated
cc-by-4.0Added 6 May 2025
MLX Community · 809MTrending
whisper-large-v3-turbo
MLX Audio
Downloads, 30 days
382K
Memory (est.)
—
Download (est.)
—
Devices
Runs on: Mac
mitAdded 1 Oct 2024
Alibaba Qwen · 7.6BTrending
Qwen2.5-Coder-7B
MLX Text
Downloads, 30 days
266K
Memory (est.)
7 GB
Download (est.)
From 4.3 GB
Devices
Runs on: Mac
apache-2.0Added 18 Sept 2024
MLX Community · 8.3B
olmOCR-2-7B-1025
MLX Text · Vision
Downloads, 30 days
248K
Memory (est.)
21 GB
Download (est.)
From 16.6 GB
Devices
Runs on: Mac
apache-2.0Added 22 Oct 2025
ResembleAI · 626M
chatterbox-turbo
MLX Audio
Downloads, 30 days
242K
Memory (est.)
3 GB
Download (est.)
From 415 MB
Devices
Runs on: Mac
mitAdded 17 Dec 2025
MLX Community · 242M
whisper-small
MLX Audio
Downloads, 30 days
205K
Memory (est.)
—
Download (est.)
—
Devices
Runs on: Mac
apache-2.0Added 8 Mar 2024
OpenAI · 21BTrending
gpt-oss-20b
MLXCore AI Text
Downloads, 30 days
179K
Memory (est.)
16 GB
Download (est.)
From 11.8 GB
Devices
Runs on: Mac
apache-2.0Added 29 Aug 2025
MLX Community · 938MTrending
Qwen3-ASR-0.6B
MLX
Downloads, 30 days
174K
Memory (est.)
3 GB
Download (est.)
From 708 MB
Devices
Runs on: Mac
apache-2.0Added 29 Jan 2026
Browse all 2,359 models

Recently added

New MLX builds on Hugging Face and new entries in Apple’s Core AI catalogue.

Full changelog
  1. Inkling-2-SmallAddedMLX
    thinkingmachines. An MLX build (4-bit).
  2. NVIDIA-Nemotron-Labs-Dory1.0-4BAddedMLX
    MLX Community. An MLX build (8-bit).
  3. Mistral-7B-Instruct-v0.3New recipeCore AI
    Mistral AI. Apple Core AI export recipe for iOS.
  4. Qwen3-1.7BNew recipeCore AI
    Alibaba Qwen. Apple Core AI export recipe for iOS.
  5. Qwen3-8BNew recipeCore AI
    Alibaba Qwen. Apple Core AI export recipe for iOS.
  6. OLMo-2-0425-1B-InstructNew recipeCore AI
    Ai2. Apple Core AI export recipe for iOS.