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 8 of 49 · clear filters

quantized-gemma-7b-it

mlx-community/quantized-gemma-7b-it
Text generation ↓ 312 · ♥ 6
Params
8.5B
Weights
≈ 32.6 GB
Mac memory
48 GB+

gemma-terms-of-use · updated 2024-02-22

MiMo-V2-Flash-4bit

from XiaomiMiMo/MiMo-V2-Flash
4-bit
Text generation · XiaomiMiMo ↓ 311 · ♥ 8
Params
309B
Weights
≈ 174 GB
Mac memory
256 GB+

mit · updated 2025-12-17

Hy-Memory-30B-A3B-8bit

from tencent/Hy-Memory-30B-A3B
8-bit
Text generation · tencent ↓ 311 · ♥ 1
Params
30B
Weights
≈ 32.0 GB
Mac memory
48 GB+

apache-2.0 · updated 2026-10-21

Qwen2.5-72B-Instruct-8bit

from Qwen/Qwen2.5-72B
8-bit
Text generation · Qwen ↓ 309 · ♥ 4
Params
73B
Weights
≈ 77.3 GB
Mac memory
96 GB+

qwen · updated 2024-09-19

Devstral-Small-2505-4bit

from mistralai/Devstral-Small-2505
4-bit
Text generation · Mistral ↓ 304 · ♥ 3
Params
24B
Weights
≈ 13.3 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-05-21

LFM2.5-1.2B-Thinking-OptiQ-4bit

from LiquidAI/LFM2.5-1.2B-Thinking
4-bit
Text generation · LiquidAI ↓ 304 · ♥ 0
Params
1.2B
Weights
≈ 659 MB
Mac memory
8 GB+

lfm1.0 · updated 2026-10-14

K2-Horizon-MoVA-36B-A4B-oQ4e

from IFM/K2-Horizon-MoVA-36B-A4B
4-bit
Text generation · IFM ↓ 303 · ♥ 3
Params
37B
Weights
≈ 21.1 GB
Mac memory
32 GB+

apache-2.0 · updated 2026-09-05

Qwen2.5-1.5B-Instruct-bf16

from Qwen/Qwen2.5-1.5B
bf16
Text generation · Qwen ↓ 298 · ♥ 0
Params
1.5B
Weights
≈ 3.1 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-09-18

Qwen3-1.7B-4bit-DWQ

from Qwen/Qwen3-1.7B
4-bit DWQ
Text generation · Qwen ↓ 296 · ♥ 0
Params
1.7B
Weights
≈ 968 MB
Mac memory
8 GB+

apache-2.0 · updated 2025-05-01

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

mlx-community/Mistral-7B-Instruct-v0.2-4bit
4-bit
Text generation ↓ 295 · ♥ 3
Params
7.2B
Weights
≈ 4.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-03-25

Llama-3.1-Nemotron-Nano-4B-v1.1-4bit

from nvidia/Llama-3.1-Nemotron-Nano-4B-v1.1
4-bit
Text generation · NVIDIA ↓ 294 · ♥ 0
Params
4.5B
Weights
≈ 2.5 GB
Mac memory
8 GB+

nvidia-open-model-license · updated 2025-06-04

Llama-Guard-3-1B-chat-mlx-4bit

from meta-llama/Llama-Guard-3-1B
4-bit
Text generation · Meta · Llama ↓ 294 · ♥ 0
Params
1.5B
Weights
≈ 843 MB
Mac memory
8 GB+

llama3.2 · updated 2026-10-14

Qwen2.5-7B-Instruct-8bit

from Qwen/Qwen2.5-7B
8-bit
Text generation · Qwen ↓ 293 · ♥ 3
Params
7.6B
Weights
≈ 8.1 GB
Mac memory
16 GB+

apache-2.0 · updated 2024-09-18

NVIDIA-Nemotron-3-Nano-4B-BF16

from nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16
bf16
Text generation · NVIDIA ↓ 292 · ♥ 2
Params
4.0B
Weights
≈ 7.9 GB
Mac memory
16 GB+

nvidia-nemotron-open-model-license · updated 2026-03-21

Qwen3-4B-Instruct-2507-bf16

from Qwen/Qwen3-4B-Instruct-2507
bf16
Text generation · Qwen ↓ 292 · ♥ 2
Params
4.0B
Weights
≈ 8.0 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-01-07

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

from lq-preview/LFM2.5-24B-A2B-Preview
5-bit
Text generation · lq-preview ↓ 292 · ♥ 0
Params
24B
Weights
≈ 16.5 GB
Mac memory
24 GB+

lfm1.0 · updated 2026-10-13

GLM-4.6-4bit

from zai-org/GLM-4.6
4-bit
Text generation · zai-org ↓ 287 · ♥ 15
Params
353B
Weights
≈ 199 GB
Mac memory
256 GB+

mit · updated 2025-09-30

K2-Horizon-0.9B-chat-mlx-16bit

from IFM/K2-Horizon-0.9B
16-bit
Text generation · IFM ↓ 287 · ♥ 0
Params
1.1B
Weights
≈ 2.2 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-10-14

Kimi-K2-Thinking-4bit

from moonshotai/Kimi-K2-Thinking
4-bit
Text generation · Moonshot · Kimi ↓ 286 · ♥ 9
Params
1.0T
Weights
≈ 578 GB
Mac memory
> 512 GB

modified-mit · updated 2025-11-06

Phi-4-mini-reasoning-4bit

from microsoft/Phi-4-mini-reasoning
4-bit
Text generation · Microsoft · Phi ↓ 286 · ♥ 0
Params
3.8B
Weights
≈ 2.2 GB
Mac memory
8 GB+

mit · updated 2025-05-01

Laguna-S-2.1-oQ2e-fast

from poolside/Laguna-S-2.1
2-bit
Text generation · poolside ↓ 283 · ♥ 13
Params
118B
Weights
≈ 36.8 GB
Mac memory
48 GB+

openmdw-1.1 · updated 2026-07-24

Meta-Llama-3-70B-Instruct-4bit

mlx-community/Meta-Llama-3-70B-Instruct-4bit
4-bit
Text generation ↓ 281 · ♥ 7
Params
71B
Weights
≈ 39.7 GB
Mac memory
48 GB+

llama3 · updated 2024-04-30

translategemma-4b-it-8bit

from google/translategemma-4b-it
8-bit
Text generation · Google · Gemma ↓ 280 · ♥ 5
Params
3.9B
Weights
≈ 4.1 GB
Mac memory
8 GB+

gemma · updated 2026-01-16

Qwen3.6-35B-A3B-MTP-5bit

from Qwen/Qwen3.6-35B-A3B
5-bit
Text generation · Qwen ↓ 278 · ♥ 6
Params
845M
Weights
≈ 581 MB
Mac memory
8 GB+

apache-2.0 · updated 2026-06-01

EuroLLM-22B-Instruct-2512-mlx-4bit

from utter-project/EuroLLM-22B-Instruct-2512
4-bit
Text generation · utter-project ↓ 278 · ♥ 3
Params
23B
Weights
≈ 12.7 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-12-16

Nemotron-Labs-Diffusion-3B-4bit

from nvidia/Nemotron-Labs-Diffusion-3B
4-bit
Text generation · NVIDIA ↓ 277 · ♥ 4
Params
3.8B
Weights
≈ 2.2 GB
Mac memory
8 GB+

nvidia-nemotron-open-model-license · updated 2026-06-27

Devstral-Small-2505-3bit

from mistralai/Devstral-Small-2505
3-bit
Text generation · Mistral ↓ 277 · ♥ 1
Params
24B
Weights
≈ 10.3 GB
Mac memory
16 GB+

apache-2.0 · updated 2025-05-21

LFM2.5-24B-A2B-Preview-nvfp4

from lq-preview/LFM2.5-24B-A2B-Preview
4-bit
Text generation · lq-preview ↓ 277 · ♥ 0
Params
24B
Weights
≈ 13.5 GB
Mac memory
24 GB+

lfm1.0 · updated 2026-10-13

DeepSeek-Coder-V2-Lite-Instruct-4bit-AWQ

from deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct
4-bit
Text generation · DeepSeek ↓ 276 · ♥ 1
Params
16B
Weights
≈ 8.8 GB
Mac memory
16 GB+

deepseek-license · updated 2025-06-27

Dolphin3.0-Llama3.1-8B-MLX-6bit

from dphn/Dolphin3.0-Llama3.1-8B
6-bit
Text generation · dphn ↓ 276 · ♥ 1
Params
8.0B
Weights
≈ 6.5 GB
Mac memory
16 GB+

llama3.1 · updated 2026-04-01

Devstral-2-123B-Instruct-2512-4bit

from mistralai/Devstral-2-123B-Instruct-2512
4-bit
Text generation · Mistral ↓ 275 · ♥ 1
Params
125B
Weights
≈ 70.3 GB
Mac memory
96 GB+

other · updated 2025-12-10

Phi-3.5-vision-instruct-4bit

mlx-community/Phi-3.5-vision-instruct-4bit
4-bit
Text generation ↓ 270 · ♥ 6
Params
4.1B
Weights
≈ 2.3 GB
Mac memory
8 GB+

mit · updated 2025-04-19

Qwen3.8-27B-MTP-mxfp4

from Qwen/Qwen3.8-27B
MXFP4
Text generation · Qwen ↓ 270 · ♥ 3
Params
425M
Weights
≈ 239 MB
Mac memory
8 GB+

apache-2.0 · updated 2026-11-11

LFM2.5-24B-A2B-Preview-6bit

from lq-preview/LFM2.5-24B-A2B-Preview
6-bit
Text generation · lq-preview ↓ 270 · ♥ 0
Params
24B
Weights
≈ 19.5 GB
Mac memory
24 GB+

lfm1.0 · updated 2026-10-13

Qwen3-0.6B-4bit-DWQ

from Qwen/Qwen3-0.6B
4-bit DWQ
Text generation · Qwen ↓ 270 · ♥ 0
Params
596M
Weights
≈ 335 MB
Mac memory
8 GB+

apache-2.0 · updated 2025-05-03

Index-Translate-9B-MLX-mxfp8

from IndexTeam/Index-Translate-9B
MXFP8
Text generation · IndexTeam ↓ 269 · ♥ 1
Params
9.0B
Weights
≈ 9.5 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-10-29

NVIDIA-Nemotron-3-Nano-30B-A3B-MLX-8Bit

from nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16
8-bit
Text generation · NVIDIA ↓ 267 · ♥ 2
Params
8.9B
Weights
≈ 10.4 GB
Mac memory
16 GB+

nvidia-open-model-license · updated 2025-12-17

Llama-Guard-3-1B-chat-mlx-8bit

from meta-llama/Llama-Guard-3-1B
8-bit
Text generation · Meta · Llama ↓ 267 · ♥ 0
Params
1.5B
Weights
≈ 1.6 GB
Mac memory
8 GB+

llama3.2 · updated 2026-10-14

Qwen2.5-7B-4bit

mlx-community/Qwen2.5-7B-4bit
4-bit
Text generation ↓ 267 · ♥ 0
Params
7.6B
Weights
≈ 4.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-09-18

bitnet-b1.58-2B-4T

from microsoft/bitnet-b1.58-2B-4T
8-bit
Text generation · Microsoft · Phi ↓ 265 · ♥ 7
Params
850M
Weights
≈ 1.2 GB
Mac memory
8 GB+

mit · updated 2025-06-10

Apriel-1.5-15b-Thinker-CYBERSEC-MERGED-4bit

from sainikhiljuluri2015/Apriel-1.5-15b-Thinker-CYBERSEC-MERGED
4-bit
Text generation · sainikhiljuluri2015 ↓ 265 · ♥ 0
Params
14B
Weights
≈ 8.1 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-10-18

Meta-Llama-3.1-8B-4bit

mlx-community/Meta-Llama-3.1-8B-4bit
4-bit
Text generation ↓ 261 · ♥ 6
Params
8.0B
Weights
≈ 4.5 GB
Mac memory
8 GB+

llama3.1 · updated 2024-07-23

Qwen2.5-Coder-32B-Instruct-3bit

from Qwen/Qwen2.5-Coder-32B-Instruct
3-bit
Text generation · Qwen ↓ 258 · ♥ 4
Params
33B
Weights
≈ 14.3 GB
Mac memory
24 GB+

apache-2.0 · updated 2024-11-23

granite-4.0-h-1b-4bit

from ibm-granite/granite-4.0-h-1b
4-bit
Text generation · IBM · Granite ↓ 258 · ♥ 1
Params
1.5B
Weights
≈ 823 MB
Mac memory
8 GB+

apache-2.0 · updated 2025-10-28

MiniMax-M2.1-3bit

from MiniMaxAI/MiniMax-M2.1
3-bit
Text generation · MiniMaxAI ↓ 257 · ♥ 5
Params
229B
Weights
≈ 100 GB
Mac memory
128 GB+

modified-mit · updated 2026-02-18

Qwen3-Coder-30B-A3B-Instruct-5bit

from Qwen/Qwen3-Coder-30B-A3B-Instruct
5-bit
Text generation · Qwen ↓ 256 · ♥ 2
Params
31B
Weights
≈ 21.0 GB
Mac memory
32 GB+

apache-2.0 · updated 2025-07-31

K2-Horizon-MoVA-36B-A4B-8bit

from IFM/K2-Horizon-MoVA-36B-A4B
8-bit
Text generation · IFM ↓ 256 · ♥ 1
Params
37B
Weights
≈ 39.8 GB
Mac memory
48 GB+

apache-2.0 · updated 2026-09-29

Qwen2.5-7B-Instruct-1M-4bit

from Qwen/Qwen2.5-7B-Instruct-1M
4-bit
Text generation · Qwen ↓ 254 · ♥ 11
Params
7.6B
Weights
≈ 4.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-01-26

exaone-4.0-1.2b-4bit

from LGAI-EXAONE/EXAONE-4.0-1.2B
4-bit
Text generation · LGAI-EXAONE ↓ 254 · ♥ 0
Params
1.3B
Weights
≈ 720 MB
Mac memory
8 GB+

exaone · updated 2025-07-15

Nemotron-Cascade-2-30B-A3B-8bit

from nvidia/Nemotron-Cascade-2-30B-A3B
8-bit
Text generation · NVIDIA ↓ 252 · ♥ 8
Params
32B
Weights
≈ 33.6 GB
Mac memory
48 GB+

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

Nemotron-3-Ultra-550B-A55B

from nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4
4-bit
Text generation · NVIDIA ↓ 252 · ♥ 5
Params
549B
Weights
≈ 313 GB
Mac memory
512 GB+

nvidia-open-model-license · updated 2026-06-05

deepseek-ai-DeepSeek-V4-Flash-4bit

from deepseek-ai/DeepSeek-V4-Flash
4-bit
Text generation · DeepSeek ↓ 252 · ♥ 5
Params
284B
Weights
≈ 160 GB
Mac memory
192 GB+

mit · updated 2026-04-24

SmolLM-360M-Instruct-4bit

mlx-community/SmolLM-360M-Instruct-4bit
4-bit
Text generation ↓ 251 · ♥ 0
Params
362M
Weights
≈ 204 MB
Mac memory
8 GB+

apache-2.0 · updated 2024-07-16

gemma-4-E4B-it-qat-assistant-bf16

from google/gemma-4-E4B-it-qat-q4_0-unquantized-assistant
bf16
Text generation · Google · Gemma ↓ 250 · ♥ 2
Params
79M
Weights
≈ 158 MB
Mac memory
8 GB+

gemma · updated 2026-06-05

Dolphin-Mistral-24B-Venice-Edition-4bit

from dphn/Dolphin-Mistral-24B-Venice-Edition
4-bit
Text generation · dphn ↓ 250 · ♥ 1
Params
24B
Weights
≈ 13.3 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-07-06

Llama-3.1-Nemotron-70B-Instruct-HF-bf16

from nvidia/Llama-3.1-Nemotron-70B-Instruct-HF
bf16
Text generation · NVIDIA ↓ 250 · ♥ 1
Params
71B
Weights
≈ 141 GB
Mac memory
192 GB+

llama3.1 · updated 2024-10-16

Llama-3.3-70B-Instruct-4bit-DWQ

from mlx-community/Llama-3.3-70B-Instruct-bf16
4-bit DWQ
Text generation ↓ 249 · ♥ 5
Params
71B
Weights
≈ 39.7 GB
Mac memory
48 GB+

llama3.3 · updated 2025-05-23