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 1h ago. Weights size and Mac memory are estimates from parameter count and quantisation.

Alert me about new Apple/MLX models

2,922 builds · page 4 of 49 · clear filters

translategemma-12b-it-8bit

from google/translategemma-12b-it
8-bit
Text generation · Google · Gemma ↓ 1.1K · ♥ 8
Params
12B
Weights
≈ 12.5 GB
Mac memory
16 GB+

gemma · updated 2026-01-16

Qwen2.5-Coder-7B-8bit

from Qwen/Qwen2.5-7B
8-bit
Text generation · Qwen ↓ 1.0K · ♥ 2
Params
7.6B
Weights
≈ 8.1 GB
Mac memory
16 GB+

apache-2.0 · updated 2024-09-18

MiniCPM5-2B-chat-mlx-4bit

from openbmb/MiniCPM5-2B
4-bit
Text generation · openbmb ↓ 1.0K · ♥ 2
Params
2.5B
Weights
≈ 1.4 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-10-14

Phi-3-mini-4k-instruct-4bit

mlx-community/Phi-3-mini-4k-instruct-4bit
4-bit
Text generation ↓ 1.0K · ♥ 12
Params
3.8B
Weights
≈ 2.1 GB
Mac memory
8 GB+

mit · updated 2024-07-11

Qwen3.5-9B-MTP-4bit

from Qwen/Qwen3.5-9B
4-bit
Text generation · Qwen ↓ 1.0K · ♥ 4
Params
243M
Weights
≈ 137 MB
Mac memory
8 GB+

apache-2.0 · updated 2026-06-01

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

from google/gemma-4-31B-it-assistant
bf16
Text generation · Google · Gemma ↓ 1.0K · ♥ 21
Params
470M
Weights
≈ 939 MB
Mac memory
8 GB+

gemma · updated 2026-05-05

Qwen2.5-Coder-7B-Instruct-bf16

from Qwen/Qwen2.5-Coder-7B
bf16
Text generation · Qwen ↓ 1.0K · ♥ 2
Params
7.6B
Weights
≈ 15.2 GB
Mac memory
24 GB+

apache-2.0 · updated 2024-09-18

Qwen2.5-72B-Instruct-4bit

from Qwen/Qwen2.5-72B-Instruct
4-bit
Text generation · Qwen ↓ 1.0K · ♥ 7
Params
73B
Weights
≈ 40.9 GB
Mac memory
48 GB+

qwen · updated 2024-09-18

Qwen3.5-4B-3D-Coder-OptiQ-4bit

from ccore/Qwen3.5-4B-3D-Coder
4-bit
Text generation · ccore ↓ 1.0K · ♥ 2
Params
4.2B
Weights
≈ 2.4 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-11-03

GLM-4.7-Flash-6bit

from zai-org/GLM-4.7-Flash
6-bit
Text generation · zai-org ↓ 991 · ♥ 7
Params
30B
Weights
≈ 24.3 GB
Mac memory
32 GB+

mit · updated 2026-01-25

Qwen3-0.6B-bf16

from Qwen/Qwen3-0.6B
bf16
Text generation · Qwen ↓ 986 · ♥ 5
Params
596M
Weights
≈ 1.2 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-04-28

OpenCodeInterpreter-SC2-3B-4bit

mlx-community/OpenCodeInterpreter-SC2-3B-4bit
4-bit
Text generation ↓ 986 · ♥ 0
Params
3.2B
Weights
≈ 2.0 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-03-03

Qwen2.5-Coder-3B-4bit

from Qwen/Qwen2.5-Coder-3B
4-bit
Text generation · Qwen ↓ 979 · ♥ 2
Params
3.1B
Weights
≈ 1.7 GB
Mac memory
8 GB+

qwen-research · updated 2024-11-12

humanizer-1B-OptiQ-4bit

from mlx-community/MiniCPM5-1B-OptiQ-4bit
4-bit
Text generation ↓ 978 · ♥ 9
Params
1.1B
Weights
≈ 608 MB
Mac memory
8 GB+

apache-2.0 · updated 2026-09-15

gemma-4-12b-coder-fable5-composer2.5-8bit

from yuxinlu1/gemma-4-12B-coder-fable5-composer2.5-v1
8-bit
Text generation · yuxinlu1 ↓ 968 · ♥ 39
Params
12B
Weights
≈ 12.7 GB
Mac memory
24 GB+

apache-2.0 · updated 2026-06-21

Qwen3-8B-8bit

from Qwen/Qwen3-8B
8-bit
Text generation · Qwen ↓ 953 · ♥ 3
Params
8.2B
Weights
≈ 8.7 GB
Mac memory
16 GB+

apache-2.0 · updated 2025-04-28

Qwen2.5-Coder-32B-Instruct-6bit

from Qwen/Qwen2.5-Coder-32B-Instruct
6-bit
Text generation · Qwen ↓ 944 · ♥ 2
Params
33B
Weights
≈ 26.6 GB
Mac memory
32 GB+

apache-2.0 · updated 2024-11-23

Josiefied-Qwen3-14B-abliterated-v3-4bit

from Goekdeniz-Guelmez/Josiefied-Qwen3-14B-abliterated-v3
4-bit
Text generation · Goekdeniz-Guelmez ↓ 943 · ♥ 1
Params
15B
Weights
≈ 8.3 GB
Mac memory
16 GB+

updated 2025-06-04

NVIDIA-Nemotron-3.5-Lightning-30B-A3B-bf16

from nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16
bf16
Text generation · NVIDIA ↓ 939 · ♥ 0
Params
32B
Weights
≈ 63.2 GB
Mac memory
96 GB+

openmdw-1.1 · updated 2026-08-11

MiniMax-M2.1-4bit

from MiniMaxAI/MiniMax-M2.1
4-bit
Text generation · MiniMaxAI ↓ 937 · ♥ 5
Params
229B
Weights
≈ 129 GB
Mac memory
192 GB+

modified-mit · updated 2026-01-15

Kimi-K2.6-mlx-DQ3_K_M-q8

from moonshotai/Kimi-K2.6
8-bit
Text generation · Moonshot · Kimi ↓ 933 · ♥ 9
Params
1.0T
Weights
≈ 1091 GB
Mac memory
> 512 GB

modified-mit · updated 2026-04-21

Qwen3-30B-A3B-8bit

from Qwen/Qwen3-30B-A3B
8-bit
Text generation · Qwen ↓ 933 · ♥ 8
Params
31B
Weights
≈ 32.4 GB
Mac memory
48 GB+

apache-2.0 · updated 2025-04-28

Ornith-1.5-9B-OptiQ-4bit

from ornith-ai/Ornith-1.5-9B-MLX
4-bit
Text generation · ornith-ai ↓ 929 · ♥ 3
Params
9.0B
Weights
≈ 5.0 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-09-15

NVIDIA-Nemotron-3.5-Lightning-30B-A3B-mxfp8

from nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16
MXFP8
Text generation · NVIDIA ↓ 919 · ♥ 0
Params
32B
Weights
≈ 33.6 GB
Mac memory
48 GB+

openmdw-1.1 · updated 2026-08-11

Qwen3-Coder-480B-A35B-Instruct-4bit

from Qwen/Qwen3-Coder-480B-A35B-Instruct
4-bit
Text generation · Qwen ↓ 914 · ♥ 19
Params
480B
Weights
≈ 270 GB
Mac memory
512 GB+

apache-2.0 · updated 2025-07-22

Meta-Llama-3.1-8B-Instruct-bf16

from meta-llama/Llama-3.1-8B-Instruct
bf16
Text generation · Meta · Llama ↓ 914 · ♥ 3
Params
8.0B
Weights
≈ 16.1 GB
Mac memory
24 GB+

llama3.1 · updated 2024-10-19

granite-4.1-8b-4bit

from ibm-granite/granite-4.1-8b
4-bit
Text generation · IBM · Granite ↓ 894 · ♥ 0
Params
8.4B
Weights
≈ 4.7 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-05-03

Qwen3-4B-Thinking-2507-4bit

from Qwen/Qwen3-4B-Thinking-2507
4-bit
Text generation · Qwen ↓ 891 · ♥ 4
Params
4.0B
Weights
≈ 2.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-08-06

gemma-3-text-4b-it-4bit

from mlx-community/gemma-3-4b-it-bf16
4-bit
Text generation ↓ 890 · ♥ 1
Params
4.6B
Weights
≈ 2.6 GB
Mac memory
8 GB+

gemma · updated 2025-03-18

DeepSeek-V4-Pro

from deepseek-ai/DeepSeek-V4-Pro
8-bit
Text generation · DeepSeek ↓ 886 · ♥ 5
Params
267B
Weights
≈ 285 GB
Mac memory
512 GB+

mit · updated 2026-06-07

Qwen1.5-MoE-A2.7B-Chat-4bit

mlx-community/Qwen1.5-MoE-A2.7B-Chat-4bit
4-bit
Text generation ↓ 884 · ♥ 7
Params
14B
Weights
≈ 8.5 GB
Mac memory
16 GB+

tongyi-qianwen · updated 2024-04-02

Qwen2.5-Coder-1.5B-Instruct-bf16

from Qwen/Qwen2.5-Coder-1.5B
bf16
Text generation · Qwen ↓ 883 · ♥ 2
Params
1.5B
Weights
≈ 3.1 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-09-18

MiniCPM5-2B-OptiQ-4bit

from openbmb/MiniCPM5-2B
4-bit
Text generation · openbmb ↓ 877 · ♥ 4
Params
2.5B
Weights
≈ 1.4 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-09-14

Qwen3.8-27B-MTP-bf16

from Qwen/Qwen3.8-27B
bf16
Text generation · Qwen ↓ 875 · ♥ 13
Params
425M
Weights
≈ 849 MB
Mac memory
8 GB+

apache-2.0 · updated 2026-08-14

Mistral-Large-4-1T-A52B-Text-3bit

from mistralai/Mistral-Large-4-1T-A52B
3-bit
Text generation · Mistral ↓ 874 · ♥ 0
Params
1.0T
Weights
≈ 459 GB
Mac memory
> 512 GB

mistral-ecosystem-open-weight-license · updated 2026-10-29

DeepSeek-V4-Flash-0731-OptiQ-2bit

from deepseek-ai/DeepSeek-V4-Flash-0731
2-bit
Text generation · DeepSeek ↓ 868 · ♥ 11
Params
304B
Weights
≈ 95.3 GB
Mac memory
128 GB+

mit · updated 2026-09-16

OLMo-2-1124-7B-Instruct-4bit

from allenai/OLMo-2-1124-7B-Instruct
4-bit
Text generation · Ai2 · OLMo ↓ 868 · ♥ 0
Params
7.3B
Weights
≈ 4.1 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-11-27

Phi-3-mini-128k-instruct-4bit

mlx-community/Phi-3-mini-128k-instruct-4bit
4-bit
Text generation ↓ 863 · ♥ 14
Params
3.8B
Weights
≈ 2.1 GB
Mac memory
8 GB+

mit · updated 2024-07-11

GLM-4-9B-0414-4bit

from zai-org/GLM-4-9B-0414
4-bit
Text generation · zai-org ↓ 852 · ♥ 2
Params
9.4B
Weights
≈ 5.3 GB
Mac memory
8 GB+

mit · updated 2025-04-17

Mistral-7B-Instruct-v0.2-4-bit

mlx-community/Mistral-7B-Instruct-v0.2-4-bit
4-bit
Text generation ↓ 849 · ♥ 24
Params
—
Weights
—
Mac memory
—

apache-2.0 · updated 2023-12-27

gemma-3-1b-it-4bit-DWQ

from google/gemma-3-1b-it
4-bit DWQ
Text generation · Google · Gemma ↓ 848 · ♥ 0
Params
1.3B
Weights
≈ 732 MB
Mac memory
8 GB+

gemma · updated 2025-05-14

MiMo-V2.6-Pro-RL-mxfp4-q8

from XiaomiMiMo/MiMo-V2.6-Pro-RL
MXFP4
Text generation · XiaomiMiMo ↓ 844 · ♥ 4
Params
1.0T
Weights
≈ 575 GB
Mac memory
> 512 GB

mit · updated 2026-09-22

gemma-4-e2b-it-qat-OptiQ-4bit

from google/gemma-4-E2B-it-qat-q4_0-unquantized
4-bit
Text generation · Google · Gemma ↓ 837 · ♥ 7
Params
4.6B
Weights
≈ 2.6 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-10-18

LFM2.5-24B-A2B-OptiQ-4bit

from LiquidAI/LFM2.5-24B-A2B
4-bit
Text generation · LiquidAI ↓ 836 · ♥ 1
Params
24B
Weights
≈ 13.5 GB
Mac memory
24 GB+

lfm1.0 · updated 2026-10-16

DeepSeek-V3.1-4bit

from deepseek-ai/DeepSeek-V3.1
4-bit
Text generation · DeepSeek ↓ 830 · ♥ 6
Params
671B
Weights
≈ 378 GB
Mac memory
512 GB+

mit · updated 2025-08-21

Mistral-Small-3.2-24B-Instruct-2506-8bit

from mistralai/Mistral-Small-3.2-24B-Instruct-2506
8-bit
Text generation · Mistral ↓ 823 · ♥ 4
Params
24B
Weights
≈ 25.0 GB
Mac memory
32 GB+

apache-2.0 · updated 2025-06-22

Qwen2.5-Coder-1.5B-Instruct-8bit

from Qwen/Qwen2.5-Coder-1.5B
8-bit
Text generation · Qwen ↓ 822 · ♥ 1
Params
1.5B
Weights
≈ 1.6 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-09-18

gemma-4-26B-A4B-it-assistant-bf16

from google/gemma-4-26B-A4B-it-assistant
bf16
Text generation · Google · Gemma ↓ 821 · ♥ 22
Params
420M
Weights
≈ 839 MB
Mac memory
8 GB+

gemma · updated 2026-05-05

Qwen3-4B-Instruct-2507-8bit

from Qwen/Qwen3-4B-Instruct-2507
8-bit
Text generation · Qwen ↓ 821 · ♥ 5
Params
4.0B
Weights
≈ 4.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-08-06

K2-Horizon-7B-oQ6e

from IFM/K2-Horizon-7B
6-bit
Text generation · IFM ↓ 816 · ♥ 5
Params
9.0B
Weights
≈ 7.3 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-09-05

DeepSeek-R1-0528-4bit

mlx-community/DeepSeek-R1-0528-4bit
4-bit
Text generation ↓ 812 · ♥ 19
Params
671B
Weights
≈ 378 GB
Mac memory
512 GB+

updated 2025-05-29

Qwen3-1.7B-4bit-DWQ-053125

from Qwen/Qwen3-1.7B
4-bit DWQ
Text generation · Qwen ↓ 812 · ♥ 2
Params
1.7B
Weights
≈ 968 MB
Mac memory
8 GB+

apache-2.0 · updated 2025-06-01

Qwen2.5-Coder-3B-Instruct-8bit

from Qwen/Qwen2.5-Coder-3B-Instruct
8-bit
Text generation · Qwen ↓ 809 · ♥ 3
Params
3.1B
Weights
≈ 3.3 GB
Mac memory
8 GB+

qwen-research · updated 2024-11-11

NVIDIA-Nemotron-3.5-Lightning-30B-A3B-OptiQ-4bit

from nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16
4-bit
Text generation · NVIDIA ↓ 807 · ♥ 5
Params
32B
Weights
≈ 17.8 GB
Mac memory
24 GB+

nvidia-open-model-license · updated 2026-09-15

MiMo-V2.6-Flash-RL-mxfp4-q8

from XiaomiMiMo/MiMo-V2.6-Flash-RL
MXFP4
Text generation · XiaomiMiMo ↓ 806 · ♥ 2
Params
309B
Weights
≈ 174 GB
Mac memory
256 GB+

mit · updated 2026-09-22

flint-oQ8e

from RecursiveSI/flint
8-bit
Text generation · RecursiveSI ↓ 804 · ♥ 2
Params
40B
Weights
≈ 42.3 GB
Mac memory
64 GB+

apache-2.0 · updated 2026-10-27

gemma-3-12b-it-qat-abliterated-lm-4bit

from mlabonne/gemma-3-12b-it-qat-abliterated
4-bit
Text generation · mlabonne ↓ 799 · ♥ 1
Params
12B
Weights
≈ 6.6 GB
Mac memory
16 GB+

gemma · updated 2025-09-29