Skip to content
AI Model Radar
Apple silicon

Apple MLX models

5,841 models converted for Apple MLX, listed from Hugging Face’s mlx-community, refreshed 2h ago. Weights size and Mac memory are estimates from parameter count and quantisation.

Alert me about new Apple/MLX models

5,841 builds · page 35 of 98

pocket-tts-4bit

from kyutai/pocket-tts
4-bit
Text to speech · kyutai ↓ 98 · ♥ 5
Params
118M
Weights
≈ 89 MB
Mac memory
8 GB+

cc-by-4.0 · updated 2026-01-20

Mellum-4b-base-4bit

from JetBrains/Mellum-4b-base
4-bit
Text generation · JetBrains ↓ 98 · ♥ 3
Params
4.0B
Weights
≈ 2.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-06-28

SongGeneration-v2-medium

from tencent/SongGeneration
Text to audio · tencent ↓ 98 · ♥ 2
Params
2.8B
Weights
≈ 5.6 GB
Mac memory
16 GB+

other · updated 2026-05-31

Qwen3-14B-bf16

from Qwen/Qwen3-14B
bf16
Text generation · Qwen ↓ 98 · ♥ 1
Params
15B
Weights
≈ 29.5 GB
Mac memory
48 GB+

apache-2.0 · updated 2025-04-28

3b-de-ft-research_release-6bit

from canopylabs/orpheus-3b-0.1-pretrained
6-bit
Text to speech · canopylabs ↓ 98 · ♥ 0
Params
3.3B
Weights
≈ 2.7 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-04-09

3b-es_it-pretrain-research_release-4bit

from meta-llama/Llama-3.2-3B-Instruct
4-bit
Text to speech · Meta · Llama ↓ 98 · ♥ 0
Params
3.3B
Weights
≈ 1.9 GB
Mac memory
8 GB+

llama3.2 · updated 2025-04-09

ACE-Step-1.5-planner-1.7B-4bit

from ACE-Step/Ace-Step1.5
4-bit
Text to audio · ACE-Step ↓ 98 · ♥ 0
Params
1.9B
Weights
≈ 1.7 GB
Mac memory
8 GB+

mit · updated 2026-10-28

MiMo-V2.5-ASR-MLX-bf16

from XiaomiMiMo/MiMo-V2.5-ASR
bf16
Speech recognition · XiaomiMiMo ↓ 98 · ♥ 0
Params
7.6B
Weights
≈ 16.0 GB
Mac memory
24 GB+

apache-2.0 · updated 2026-05-30

Nanonets-OCR2-3B-4bit

from Qwen/Qwen2.5-VL-3B-Instruct
4-bit
Vision + text · Qwen ↓ 98 · ♥ 0
Params
3.8B
Weights
≈ 3.1 GB
Mac memory
8 GB+

updated 2025-10-14

Qwen2.5-Coder-0.5B-8bit

from Qwen/Qwen2.5-Coder-0.5B
8-bit
Text generation · Qwen ↓ 98 · ♥ 0
Params
494M
Weights
≈ 525 MB
Mac memory
8 GB+

apache-2.0 · updated 2024-11-11

Qwen3-ASR-0.6B-5bit

mlx-community/Qwen3-ASR-0.6B-5bit
5-bit
Uncategorised ↓ 98 · ♥ 0
Params
782M
Weights
≈ 783 MB
Mac memory
8 GB+

apache-2.0 · updated 2026-01-29

llm-jp-3-7.2b-instruct3

from llm-jp/llm-jp-3-7.2b-instruct3
Text generation · llm-jp ↓ 98 · ♥ 0
Params
7.3B
Weights
≈ 14.6 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-02-05

sqlcoder-70b-alpha-4bit-mlx

mlx-community/sqlcoder-70b-alpha-4bit-mlx
4-bit
Text generation ↓ 97 · ♥ 6
Params
—
Weights
—
Mac memory
—

cc-by-sa-4.0 · updated 2024-02-01

Qwen3.6-27B-mxfp4

from Qwen/Qwen3.6-27B
MXFP4
Vision + text · Qwen ↓ 97 · ♥ 5
Params
27B
Weights
≈ 16.1 GB
Mac memory
24 GB+

apache-2.0 · updated 2026-04-22

OpenHermes-2.5-Mistral-7B-4bit-mlx

from mistralai/Mistral-7B-v0.1
4-bit
Uncategorised · Mistral ↓ 97 · ♥ 3
Params
7.2B
Weights
≈ 4.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-01-25

VGGT-Omega-1B-512-bf16

from facebook/VGGT-Omega
bf16
Image to 3D · facebook ↓ 97 · ♥ 2
Params
1.1B
Weights
≈ 2.3 GB
Mac memory
8 GB+

fair-noncommercial-research-license · updated 2026-09-28

aya-expanse-32b-8bit

from CohereLabs/aya-expanse-32b
8-bit
Text generation · Cohere ↓ 97 · ♥ 2
Params
32B
Weights
≈ 34.3 GB
Mac memory
48 GB+

cc-by-nc-4.0 · updated 2024-10-25

3b-fr-pretrain-research_release-4bit

from meta-llama/Llama-3.2-3B-Instruct
4-bit
Text to speech · Meta · Llama ↓ 97 · ♥ 1
Params
3.3B
Weights
≈ 1.9 GB
Mac memory
8 GB+

llama3.2 · updated 2025-04-09

Josiefied-Qwen2.5-14B-Instruct-abliterated-v4-2-bit

from Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4
2-bit
Text generation · Goekdeniz-Guelmez ↓ 97 · ♥ 1
Params
15B
Weights
≈ 4.6 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-12-09

Skyfall-31B-v4.2-heretic-mlx-fp16

from coder3101/Skyfall-31B-v4.2-heretic
fp16
Uncategorised · coder3101 ↓ 97 · ♥ 1
Params
31B
Weights
≈ 62.7 GB
Mac memory
96 GB+

updated 2026-08-06

gemma-4-31B-it-qat-5bit

mlx-community/gemma-4-31B-it-qat-5bit
5-bit
Vision + text ↓ 97 · ♥ 1
Params
31B
Weights
≈ 22.2 GB
Mac memory
32 GB+

updated 2026-06-05

MinerU2.5-2509-1.2B-bf16

mlx-community/MinerU2.5-2509-1.2B-bf16
bf16
Vision + text ↓ 97 · ♥ 0
Params
1.2B
Weights
≈ 2.3 GB
Mac memory
8 GB+

agpl-3.0 · updated 2025-10-04

Olmo-3.1-32B-Instruct-8bit

from allenai/Olmo-3.1-32B-Instruct
8-bit
Text generation · Ai2 · OLMo ↓ 97 · ♥ 0
Params
32B
Weights
≈ 34.2 GB
Mac memory
48 GB+

apache-2.0 · updated 2025-12-23

Qwen3.5-4B-mixed_3_4

mlx-community/Qwen3.5-4B-mixed_3_4
4-bit
Uncategorised ↓ 97 · ♥ 0
Params
892M
Weights
≈ 1.2 GB
Mac memory
8 GB+

updated 2026-03-02

CodeLlama-70b-hf-4bit-MLX

mlx-community/CodeLlama-70b-hf-4bit-MLX
4-bit
Text generation ↓ 96 · ♥ 12
Params
—
Weights
—
Mac memory
—

llama2 · updated 2024-01-31

gte-Qwen2-7B-instruct-4bit-DWQ

from Alibaba-NLP/gte-Qwen2-7B-instruct
4-bit DWQ
Text generation · Alibaba-NLP ↓ 96 · ♥ 3
Params
7.6B
Weights
≈ 4.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-05-07

gemma-3-12b-it-4bit-DWQ

from google/gemma-3-12b-it
4-bit DWQ
Text generation · Google · Gemma ↓ 96 · ♥ 2
Params
13B
Weights
≈ 7.2 GB
Mac memory
16 GB+

gemma · updated 2025-05-18

Kimi-Linear-48B-A3B-Instruct-5bit

from moonshotai/Kimi-Linear-48B-A3B-Instruct
5-bit
Text generation · Moonshot · Kimi ↓ 96 · ♥ 1
Params
49B
Weights
≈ 33.8 GB
Mac memory
48 GB+

mit · updated 2026-01-22

Mistral-Small-3.2-24B-Instruct-2506-bf16

from mistralai/Mistral-Small-3.2-24B-Instruct-2506
bf16
Text generation · Mistral ↓ 96 · ♥ 1
Params
24B
Weights
≈ 47.1 GB
Mac memory
64 GB+

apache-2.0 · updated 2025-06-21

cogito-v2-preview-deepseek-671B-MoE-4bit

from deepcogito/cogito-v2-preview-deepseek-671B-MoE
4-bit
Text generation · deepcogito ↓ 96 · ♥ 1
Params
671B
Weights
≈ 378 GB
Mac memory
512 GB+

mit · updated 2025-08-02

Ministral-3-3B-Reasoning-2512-8bit

from mistralai/Ministral-3-3B-Base-2512
8-bit
Uncategorised · Mistral ↓ 96 · ♥ 0
Params
3.8B
Weights
≈ 4.5 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-12-03

OpenResearcher-30B-A3B-6bit

from OpenResearcher/OpenResearcher-30B-A3B
6-bit
Text generation · OpenResearcher ↓ 96 · ♥ 0
Params
32B
Weights
≈ 25.7 GB
Mac memory
32 GB+

mit · updated 2026-02-10

Qwen2.5-Coder-0.5B-Instruct-8bit

from Qwen/Qwen2.5-Coder-0.5B-Instruct
8-bit
Text generation · Qwen ↓ 96 · ♥ 0
Params
494M
Weights
≈ 525 MB
Mac memory
8 GB+

apache-2.0 · updated 2024-11-11

bark-small

mlx-community/bark-small
Text to speech ↓ 96 · ♥ 0
Params
420M
Weights
≈ 1.6 GB
Mac memory
8 GB+

mit · updated 2025-03-19

bigcode-starcoder2-15b-8bit

from bigcode/starcoder2-15b
8-bit
Text generation · BigCode ↓ 96 · ♥ 0
Params
16B
Weights
≈ 17.0 GB
Mac memory
24 GB+

bigcode-openrail-m · updated 2025-02-16

starcoder2-15b-instruct-v0.1

from bigcode/starcoder2-15b
Text generation · BigCode ↓ 96 · ♥ 0
Params
16B
Weights
≈ 31.9 GB
Mac memory
48 GB+

bigcode-openrail-m · updated 2024-04-30

LongCat-Next-4bit

from meituan-longcat/LongCat-Next
4-bit
Text generation · meituan-longcat ↓ 95 · ♥ 7
Params
69B
Weights
≈ 38.6 GB
Mac memory
48 GB+

mit · updated 2026-03-26

Laguna-S-2.1-oQ5e

from poolside/Laguna-S-2.1
5-bit
Text generation · poolside ↓ 95 · ♥ 4
Params
118B
Weights
≈ 80.9 GB
Mac memory
96 GB+

openmdw-1.1 · updated 2026-07-24

DeepSeek-R1-Distill-Qwen-7B-4bit-mlx

from deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
4-bit
Uncategorised · DeepSeek ↓ 95 · ♥ 3
Params
7.6B
Weights
≈ 4.3 GB
Mac memory
8 GB+

updated 2025-02-26

FLUX.2-klein-9B

mlx-community/FLUX.2-klein-9B
Image to image ↓ 95 · ♥ 3
Params
9.1B
Weights
≈ 18.2 GB
Mac memory
24 GB+

flux-non-commercial-license · updated 2026-02-21

DeepSeek-R1-0528-Qwen3-8B-6bit

from deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
6-bit
Text generation · DeepSeek ↓ 95 · ♥ 1
Params
8.2B
Weights
≈ 6.7 GB
Mac memory
16 GB+

mit · updated 2025-06-06

DeepSeek-V3.2-Speciale-4bit

from deepseek-ai/DeepSeek-V3.2-Speciale
4-bit
Text generation · DeepSeek ↓ 95 · ♥ 1
Params
672B
Weights
≈ 378 GB
Mac memory
512 GB+

mit · updated 2025-12-03

Qwen2.5-32B-Instruct-4bit-AWQ

from Qwen/Qwen2.5-32B-Instruct
4-bit
Text generation · Qwen ↓ 95 · ♥ 1
Params
33B
Weights
≈ 18.4 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-04-26

deepseek-vl-1.3b-chat-8bit

mlx-community/deepseek-vl-1.3b-chat-8bit
8-bit
Vision + text ↓ 95 · ♥ 1
Params
2.0B
Weights
≈ 2.1 GB
Mac memory
8 GB+

deepseek · updated 2024-12-26

CodeLlama-13b-Python-4bit-MLX

mlx-community/CodeLlama-13b-Python-4bit-MLX
4-bit
Text generation ↓ 95 · ♥ 0
Params
—
Weights
—
Mac memory
—

llama2 · updated 2024-01-29

Devstral-2-123B-Instruct-2512-bf16

from mistralai/Devstral-2-123B-Instruct-2512
bf16
Text generation · Mistral ↓ 95 · ♥ 0
Params
125B
Weights
≈ 250 GB
Mac memory
512 GB+

other · updated 2025-12-10

InternVL3_5-30B-A3B-4bit

from OpenGVLab/InternVL3_5-30B-A3B-MPO
4-bit
Vision + text · OpenGVLab ↓ 95 · ♥ 0
Params
—
Weights
—
Mac memory
—

apache-2.0 · updated 2025-10-04

XVERSE-7B-Chat-4bit

from xverse/XVERSE-7B-Chat
4-bit
Text generation · xverse ↓ 95 · ♥ 0
Params
7.3B
Weights
≈ 4.1 GB
Mac memory
8 GB+

xverse-model-license · updated 2026-09-07

embeddinggemma-300m-5bit

mlx-community/embeddinggemma-300m-5bit
5-bit
Sentence similarity ↓ 95 · ♥ 0
Params
308M
Weights
≈ 212 MB
Mac memory
8 GB+

gemma · updated 2025-09-04

granite-4.0-h-small-4bit

from ibm-granite/granite-4.0-h-small
4-bit
Text generation · IBM · Granite ↓ 95 · ♥ 0
Params
32B
Weights
≈ 18.1 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-10-02

shisa-v2.1-llama3.3-70b-mlx-8bit

from shisa-ai/shisa-v2.1-llama3.3-70b
8-bit
Text generation · shisa-ai ↓ 95 · ♥ 0
Params
71B
Weights
≈ 75.0 GB
Mac memory
96 GB+

llama3.3 · updated 2025-12-16

LFM2.5-Audio-1.5B-bf16

from LiquidAI/LFM2-1.2B
bf16
Audio to audio · LiquidAI ↓ 94 · ♥ 15
Params
1.5B
Weights
≈ 2.9 GB
Mac memory
8 GB+

lfm1.0 · updated 2026-01-15

gemma-4-26b-a4b-mxfp4

from google/gemma-4-26b-a4b
MXFP4
Vision + text · Google · Gemma ↓ 94 · ♥ 8
Params
26B
Weights
≈ 15.3 GB
Mac memory
24 GB+

apache-2.0 · updated 2026-04-03

Josiefied-Qwen2.5-7B-Instruct-abliterated-v2-2-bit

from Goekdeniz-Guelmez/Josiefied-Qwen2.5-7B-Instruct-abliterated-v2
2-bit
Text generation · Goekdeniz-Guelmez ↓ 94 · ♥ 1
Params
7.6B
Weights
≈ 2.4 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-12-08

c4ai-command-a-03-2025-4bit

from CohereLabs/c4ai-command-a-03-2025
4-bit
Text generation · Cohere ↓ 94 · ♥ 1
Params
111B
Weights
≈ 62.5 GB
Mac memory
96 GB+

cc-by-nc-4.0 · updated 2025-03-18

gemma-4-31B-it-qat-assistant-bf16

from google/gemma-4-31B-it-qat-q4_0-unquantized-assistant
bf16
Text generation · Google · Gemma ↓ 94 · ♥ 1
Params
470M
Weights
≈ 939 MB
Mac memory
8 GB+

gemma · updated 2026-06-05

gemma-4-e4b-it-5bit

from google/gemma-4-E4B-it
5-bit
Vision + text · Google · Gemma ↓ 94 · ♥ 1
Params
7.9B
Weights
≈ 6.1 GB
Mac memory
16 GB+

gemma · updated 2026-07-06