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 5 of 49 · clear filters

GLM-4.5-Air-4bit

from zai-org/GLM-4.5-Air
4-bit
Text generation · zai-org ↓ 793 · ♥ 30
Params
107B
Weights
≈ 60.1 GB
Mac memory
96 GB+

mit · updated 2025-07-28

S1-mini-MLX-4bit

from superwhisper/s1-mini
4-bit
Text generation · superwhisper ↓ 785 · ♥ 0
Params
596M
Weights
≈ 335 MB
Mac memory
8 GB+

s1-mini-license · updated 2026-08-20

phi-4-8bit

from microsoft/phi-4
8-bit
Text generation · Microsoft · Phi ↓ 777 · ♥ 14
Params
15B
Weights
≈ 15.6 GB
Mac memory
24 GB+

mit · updated 2025-01-12

Phi-3-mini-4k-instruct-4bit-no-q-embed

mlx-community/Phi-3-mini-4k-instruct-4bit-no-q-embed
4-bit
Text generation ↓ 775 · ♥ 3
Params
3.8B
Weights
≈ 2.3 GB
Mac memory
8 GB+

mit · updated 2024-04-30

Olmo-3-7B-Instruct-4bit

from allenai/Olmo-3-7B-Instruct
4-bit
Text generation · Ai2 · OLMo ↓ 768 · ♥ 2
Params
7.3B
Weights
≈ 4.1 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-11-20

granite-3.3-2b-instruct-4bit

from ibm-granite/granite-3.3-2b-instruct
4-bit
Text generation · IBM · Granite ↓ 768 · ♥ 1
Params
2.5B
Weights
≈ 1.4 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-04-16

Dolphin3.0-Llama3.1-8B-MLX-4bit

from dphn/Dolphin3.0-Llama3.1-8B
4-bit
Text generation · dphn ↓ 766 · ♥ 1
Params
8.0B
Weights
≈ 4.5 GB
Mac memory
8 GB+

llama3.1 · updated 2026-04-01

Llama-3.1-8B-Instruct

from meta-llama/Llama-3.1-8B-Instruct
Text generation · Meta · Llama ↓ 762 · ♥ 5
Params
8.0B
Weights
≈ 16.1 GB
Mac memory
24 GB+

llama3.1 · updated 2024-09-28

DeepSeek-R1-Distill-Qwen-7B-chat-mlx-16bit

from deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
16-bit
Text generation · DeepSeek ↓ 762 · ♥ 0
Params
7.6B
Weights
≈ 15.2 GB
Mac memory
24 GB+

mit · updated 2026-10-14

gemma-2-27b-it-4bit

mlx-community/gemma-2-27b-it-4bit
4-bit
Text generation ↓ 760 · ♥ 9
Params
27B
Weights
≈ 15.3 GB
Mac memory
24 GB+

gemma · updated 2024-11-06

Phi-4-mini-instruct-8bit

from microsoft/Phi-4-mini-instruct
8-bit
Text generation · Microsoft · Phi ↓ 758 · ♥ 7
Params
3.8B
Weights
≈ 4.1 GB
Mac memory
8 GB+

mit · updated 2025-03-05

Qwen3-1.7B-8bit

from Qwen/Qwen3-1.7B
8-bit
Text generation · Qwen ↓ 758 · ♥ 1
Params
1.7B
Weights
≈ 1.8 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-04-28

gemma-4-31B-it-OptiQ-4bit

from google/gemma-4-31B-it
4-bit
Text generation · Google · Gemma ↓ 757 · ♥ 15
Params
31B
Weights
≈ 18.4 GB
Mac memory
24 GB+

gemma · updated 2026-10-18

gemma-3-270m-it-8bit

from google/gemma-3-270m-it
8-bit
Text generation · Google · Gemma ↓ 752 · ♥ 2
Params
436M
Weights
≈ 463 MB
Mac memory
8 GB+

gemma · updated 2025-08-09

Qwen2.5-Coder-14B-4bit

from Qwen/Qwen2.5-Coder-14B
4-bit
Text generation · Qwen ↓ 750 · ♥ 0
Params
15B
Weights
≈ 8.3 GB
Mac memory
16 GB+

apache-2.0 · updated 2024-11-12

GLM-4.7-Flash-5bit

from zai-org/GLM-4.7-Flash
5-bit
Text generation · zai-org ↓ 745 · ♥ 0
Params
30B
Weights
≈ 20.6 GB
Mac memory
32 GB+

mit · updated 2026-01-25

Qwen3-Coder-30B-A3B-Instruct-4bit-DWQ

from Qwen/Qwen3-Coder-30B-A3B-Instruct
4-bit DWQ
Text generation · Qwen ↓ 739 · ♥ 6
Params
31B
Weights
≈ 17.2 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-08-01

Qwen2.5.1-Coder-7B-Instruct-4bit

from Qwen/Qwen2.5-Coder-7B-Instruct
4-bit
Text generation · Qwen ↓ 738 · ♥ 9
Params
7.6B
Weights
≈ 4.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-11-09

Josiefied-Qwen3-30B-A3B-abliterated-v2-4bit

from Goekdeniz-Guelmez/Josiefied-Qwen3-30B-A3B-abliterated-v2
4-bit
Text generation · Goekdeniz-Guelmez ↓ 738 · ♥ 3
Params
31B
Weights
≈ 17.2 GB
Mac memory
24 GB+

updated 2025-06-19

Kimi-K2-Instruct-0905-mlx-DQ3_K_M

from moonshotai/Kimi-K2-Instruct-0905
4-bit
Text generation · Moonshot · Kimi ↓ 735 · ♥ 12
Params
1.0T
Weights
≈ 578 GB
Mac memory
> 512 GB

modified-mit · updated 2025-11-08

granite-4.0-h-350m-8bit

from ibm-granite/granite-4.0-h-350m
8-bit
Text generation · IBM · Granite ↓ 730 · ♥ 0
Params
340M
Weights
≈ 362 MB
Mac memory
8 GB+

apache-2.0 · updated 2025-10-28

GLM-5.2-DQ4plus-q8

from zai-org/GLM-5.2
8-bit
Text generation · zai-org ↓ 724 · ♥ 18
Params
743B
Weights
≈ 790 GB
Mac memory
> 512 GB

mit · updated 2026-07-12

NVIDIA-Nemotron-3-Nano-4B-4bit

from nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16
4-bit
Text generation · NVIDIA ↓ 722 · ♥ 2
Params
4.0B
Weights
≈ 2.2 GB
Mac memory
8 GB+

nvidia-nemotron-open-model-license · updated 2026-03-20

LFM2.5-24B-A2B-Preview-bf16

from lq-preview/LFM2.5-24B-A2B-Preview
bf16
Text generation · lq-preview ↓ 722 · ♥ 1
Params
24B
Weights
≈ 47.9 GB
Mac memory
64 GB+

lfm1.0 · updated 2026-10-13

Jan-v3-4B-base-instruct-4bit

from janhq/Jan-v3-4B-base-instruct
4-bit
Text generation · janhq ↓ 716 · ♥ 2
Params
4.0B
Weights
≈ 2.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-01-27

Josiefied-Qwen2.5-3B-Instruct-abliterated-v1-4-bit

from Goekdeniz-Guelmez/Josiefied-Qwen2.5-3B-Instruct-abliterated-v1
4-bit
Text generation · Goekdeniz-Guelmez ↓ 716 · ♥ 1
Params
3.1B
Weights
≈ 1.7 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-12-20

c4ai-command-r-plus-4bit

mlx-community/c4ai-command-r-plus-4bit
4-bit
Text generation ↓ 711 · ♥ 51
Params
104B
Weights
≈ 62.9 GB
Mac memory
96 GB+

cc-by-nc-4.0 · updated 2024-04-10

gemma-4-26B-A4B-it-qat-assistant-4bit

from google/gemma-4-26B-A4B-it-qat-q4_0-unquantized-assistant
4-bit
Text generation · Google · Gemma ↓ 711 · ♥ 2
Params
420M
Weights
≈ 236 MB
Mac memory
8 GB+

gemma · updated 2026-06-05

Phi-3.5-MoE-instruct-4bit

mlx-community/Phi-3.5-MoE-instruct-4bit
4-bit
Text generation ↓ 699 · ♥ 4
Params
42B
Weights
≈ 23.6 GB
Mac memory
32 GB+

mit · updated 2024-08-24

translategemma-27b-it-4bit

from google/translategemma-27b-it
4-bit
Text generation · Google · Gemma ↓ 698 · ♥ 4
Params
27B
Weights
≈ 15.2 GB
Mac memory
24 GB+

gemma · updated 2026-01-16

Meta-Llama-3.1-70B-Instruct-8bit

mlx-community/Meta-Llama-3.1-70B-Instruct-8bit
8-bit
Text generation ↓ 697 · ♥ 4
Params
71B
Weights
≈ 75.0 GB
Mac memory
96 GB+

llama3.1 · updated 2024-11-26

NVIDIA-Nemotron-3-Nano-30B-A3B-4bit

from nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16
4-bit
Text generation · NVIDIA ↓ 691 · ♥ 4
Params
4.9B
Weights
≈ 4.2 GB
Mac memory
8 GB+

nvidia-open-model-license · updated 2025-12-31

LFM2.5-1.2B-Thinking-6bit

from LiquidAI/LFM2.5-1.2B-Thinking
6-bit
Text generation · LiquidAI ↓ 690 · ♥ 1
Params
1.2B
Weights
≈ 951 MB
Mac memory
8 GB+

lfm1.0 · updated 2026-08-13

gpt-oss-120b-mxfp4-bf16

from openai/gpt-oss-120b
MXFP4
Text generation · OpenAI · gpt-oss ↓ 685 · ♥ 3
Params
117B
Weights
≈ 68.8 GB
Mac memory
96 GB+

apache-2.0 · updated 2026-04-30

Qwen3-8B-abliterated-v2-mxfp4

from huihui-ai/Huihui-Qwen3-8B-abliterated-v2
MXFP4
Text generation · huihui-ai ↓ 684 · ♥ 1
Params
8.2B
Weights
≈ 4.6 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-12-14

Qwen2.5-Coder-14B-Instruct-abliterated-8bit

from huihui-ai/Qwen2.5-Coder-14B-Instruct-abliterated
8-bit
Text generation · huihui-ai ↓ 682 · ♥ 1
Params
15B
Weights
≈ 15.7 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-02-22

SmolLM2-360M-Instruct

from HuggingFaceTB/SmolLM2-360M-Instruct
Text generation · Hugging Face ↓ 681 · ♥ 2
Params
362M
Weights
≈ 724 MB
Mac memory
8 GB+

apache-2.0 · updated 2024-10-31

Llama-3.3-70B-Instruct-6bit

from meta-llama/Llama-3.3-70B-Instruct
6-bit
Text generation · Meta · Llama ↓ 678 · ♥ 5
Params
71B
Weights
≈ 57.3 GB
Mac memory
96 GB+

llama3.3 · updated 2024-12-06

granite-4.1-30b-4bit

from ibm-granite/granite-4.1-30b
4-bit
Text generation · IBM · Granite ↓ 675 · ♥ 2
Params
29B
Weights
≈ 16.2 GB
Mac memory
24 GB+

apache-2.0 · updated 2026-05-01

DeepSeek-V4-Flash

from deepseek-ai/DeepSeek-V4-Flash
8-bit
Text generation · DeepSeek ↓ 673 · ♥ 3
Params
51B
Weights
≈ 53.9 GB
Mac memory
64 GB+

mit · updated 2026-06-07

QED-Nano-8bit

from lm-provers/QED-Nano
8-bit
Text generation · lm-provers ↓ 668 · ♥ 0
Params
4.0B
Weights
≈ 4.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-04-02

Qwen3-1.7B-bf16

from Qwen/Qwen3-1.7B
bf16
Text generation · Qwen ↓ 663 · ♥ 0
Params
1.7B
Weights
≈ 3.4 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-04-28

gpt2-base-mlx

from openai-community/gpt2
Text generation · openai-community ↓ 658 · ♥ 0
Params
124M
Weights
≈ 498 MB
Mac memory
8 GB+

mit · updated 2025-10-09

Qwen3-Next-80B-A3B-Instruct-8bit

from Qwen/Qwen3-Next-80B-A3B-Instruct
8-bit
Text generation · Qwen ↓ 657 · ♥ 11
Params
80B
Weights
≈ 84.7 GB
Mac memory
128 GB+

apache-2.0 · updated 2025-09-13

DeepSeek-V3.1-8bit

from deepseek-ai/DeepSeek-V3.1
8-bit
Text generation · DeepSeek ↓ 647 · ♥ 3
Params
671B
Weights
≈ 713 GB
Mac memory
> 512 GB

mit · updated 2025-08-22

deepseek-ai-DeepSeek-V4-Flash-8bit

from deepseek-ai/DeepSeek-V4-Flash
8-bit
Text generation · DeepSeek ↓ 642 · ♥ 14
Params
284B
Weights
≈ 302 GB
Mac memory
512 GB+

mit · updated 2026-04-24

Qwen3-14B-8bit

from Qwen/Qwen3-14B
8-bit
Text generation · Qwen ↓ 642 · ♥ 0
Params
15B
Weights
≈ 15.7 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-04-28

Meta-Llama-3.1-405B-4bit

mlx-community/Meta-Llama-3.1-405B-4bit
4-bit
Text generation ↓ 636 · ♥ 4
Params
410B
Weights
≈ 231 GB
Mac memory
512 GB+

llama3.1 · updated 2024-07-31

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

mlx-community/Phi-3-vision-128k-instruct-4bit
4-bit
Text generation ↓ 632 · ♥ 8
Params
4.1B
Weights
≈ 2.3 GB
Mac memory
8 GB+

mit · updated 2024-06-24

GLM-4.7-4bit

from zai-org/GLM-4.7
4-bit
Text generation · zai-org ↓ 629 · ♥ 5
Params
353B
Weights
≈ 199 GB
Mac memory
256 GB+

mit · updated 2026-01-01

Nemotron-Cascade-2-30B-A3B-4bit

from nvidia/Nemotron-Cascade-2-30B-A3B
4-bit
Text generation · NVIDIA ↓ 628 · ♥ 22
Params
32B
Weights
≈ 17.8 GB
Mac memory
24 GB+

nvidia-open-model-license · updated 2026-03-20

Kimi-Linear-48B-A3B-Instruct-4bit

from moonshotai/Kimi-Linear-48B-A3B-Instruct
4-bit
Text generation · Moonshot · Kimi ↓ 627 · ♥ 11
Params
49B
Weights
≈ 27.6 GB
Mac memory
48 GB+

mit · updated 2026-01-22

Qwen3-Next-80B-A3B-Thinking-4bit

from Qwen/Qwen3-Next-80B-A3B-Thinking
4-bit
Text generation · Qwen ↓ 626 · ♥ 4
Params
80B
Weights
≈ 44.8 GB
Mac memory
64 GB+

apache-2.0 · updated 2025-09-13

DeepSeek-R1-Distill-Qwen-7B

from deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
Text generation · DeepSeek ↓ 622 · ♥ 6
Params
7.6B
Weights
≈ 15.2 GB
Mac memory
24 GB+

updated 2025-02-26

Qwen3.5-35B-A3B-OptiQ-4bit

from Qwen/Qwen3.5-35B-A3B
4-bit
Text generation · Qwen ↓ 621 · ♥ 4
Params
35B
Weights
≈ 20.4 GB
Mac memory
32 GB+

apache-2.0 · updated 2026-10-30

functiongemma-270m-it-4bit

from google/functiongemma-270m-it
4-bit
Text generation · Google · Gemma ↓ 619 · ♥ 3
Params
268M
Weights
≈ 151 MB
Mac memory
8 GB+

gemma · updated 2025-12-18

Falcon3-7B-Instruct-4bit

from tiiuae/Falcon3-7B-Instruct
4-bit
Text generation · TII · Falcon ↓ 619 · ♥ 0
Params
7.5B
Weights
≈ 4.2 GB
Mac memory
8 GB+

falcon-llm-license · updated 2024-12-17

llm-jp-4-8b-thinking-4bit

from llm-jp/llm-jp-4-8b-thinking
4-bit
Text generation · llm-jp ↓ 616 · ♥ 2
Params
8.6B
Weights
≈ 4.8 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-04-12