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

2,922 builds · page 11 of 49 · clear filters

AI21-Jamba-Reasoning-3B-4bit

from ai21labs/AI21-Jamba-Reasoning-3B
4-bit
Text generation · AI21 · Jamba ↓ 186 · ♥ 1
Params
3.0B
Weights
≈ 1.7 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-10-14

gte-Qwen2-1.5B-instruct-4bit-dwq

from Alibaba-NLP/gte-Qwen2-1.5B-instruct
4-bit DWQ
Text generation · Alibaba-NLP ↓ 186 · ♥ 1
Params
1.8B
Weights
≈ 999 MB
Mac memory
8 GB+

apache-2.0 · updated 2025-05-17

bigcode-starcoder2-15b-fp32

from bigcode/starcoder2-15b
fp32
Text generation · BigCode ↓ 186 · ♥ 0
Params
16B
Weights
≈ 63.8 GB
Mac memory
96 GB+

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

Mellum2-12B-A2.5B-Instruct-mxfp4

from JetBrains/Mellum2-12B-A2.5B-Instruct
MXFP4
Text generation · JetBrains ↓ 185 · ♥ 4
Params
12B
Weights
≈ 6.8 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-08-11

Qwen3-30B-A3B-3bit

from Qwen/Qwen3-30B-A3B
3-bit
Text generation · Qwen ↓ 185 · ♥ 2
Params
31B
Weights
≈ 13.4 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-07-01

Qwen3-4B-Instruct-2507-nvfp4

from Qwen/Qwen3-4B-Instruct-2507
4-bit
Text generation · Qwen ↓ 185 · ♥ 0
Params
4.0B
Weights
≈ 2.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-02-08

LFM2-350M-4bit

from LiquidAI/LFM2-350M
4-bit
Text generation · LiquidAI ↓ 184 · ♥ 4
Params
354M
Weights
≈ 199 MB
Mac memory
8 GB+

lfm1.0 · updated 2025-07-11

c4ai-command-r7b-12-2024-bf16

from CohereLabs/c4ai-command-r7b-12-2024
bf16
Text generation · Cohere ↓ 184 · ♥ 4
Params
8.0B
Weights
≈ 16.1 GB
Mac memory
24 GB+

cc-by-nc-4.0 · updated 2024-12-14

Qwen3-Coder-Next-mxfp4

from Qwen/Qwen3-Coder-Next
MXFP4
Text generation · Qwen ↓ 184 · ♥ 2
Params
80B
Weights
≈ 44.8 GB
Mac memory
64 GB+

apache-2.0 · updated 2026-05-06

GLM-4.7-Flash-abliterated-mxfp8

from huihui-ai/Huihui-GLM-4.7-Flash-abliterated
MXFP8
Text generation · huihui-ai ↓ 183 · ♥ 2
Params
30B
Weights
≈ 31.8 GB
Mac memory
48 GB+

mit · updated 2026-03-07

phi-4-6bit

from microsoft/phi-4
6-bit
Text generation · Microsoft · Phi ↓ 182 · ♥ 7
Params
15B
Weights
≈ 11.9 GB
Mac memory
16 GB+

mit · updated 2025-01-12

Qwen3-14B-4bit-AWQ

from unsloth/Qwen3-14B
4-bit
Text generation · unsloth ↓ 182 · ♥ 5
Params
15B
Weights
≈ 8.3 GB
Mac memory
16 GB+

apache-2.0 · updated 2025-08-29

granite-4.0-1b-4bit

from ibm-granite/granite-4.0-1b
4-bit
Text generation · IBM · Granite ↓ 182 · ♥ 1
Params
1.6B
Weights
≈ 918 MB
Mac memory
8 GB+

apache-2.0 · updated 2025-10-28

GLM-4.7-Flash-4bit-mxfp4

from zai-org/GLM-4.7-Flash
MXFP4
Text generation · zai-org ↓ 181 · ♥ 4
Params
30B
Weights
≈ 16.9 GB
Mac memory
24 GB+

mit · updated 2026-02-10

NVIDIA-Nemotron-3-Super-120B-A12B-6bit

from nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-BF16
6-bit
Text generation · NVIDIA ↓ 181 · ♥ 1
Params
121B
Weights
≈ 98.1 GB
Mac memory
128 GB+

nvidia-nemotron-open-model-license · updated 2026-03-30

IQuest-Coder-V1-40B-Loop-Instruct-4bit

from IQuestLab/IQuest-Coder-V1-40B-Loop-Instruct
4-bit
Text generation · IQuestLab ↓ 180 · ♥ 11
Params
40B
Weights
≈ 22.4 GB
Mac memory
32 GB+

iquestcoder · updated 2026-01-07

Meta-Llama-3.1-70B-Instruct-bf16

from meta-llama/Llama-3.1-70B-Instruct
bf16
Text generation · Meta · Llama ↓ 180 · ♥ 4
Params
71B
Weights
≈ 141 GB
Mac memory
192 GB+

llama3.1 · updated 2024-10-06

Devstral-Small-2505-bf16

from mistralai/Devstral-Small-2505
bf16
Text generation · Mistral ↓ 180 · ♥ 1
Params
24B
Weights
≈ 47.1 GB
Mac memory
64 GB+

apache-2.0 · updated 2025-05-21

Qwen3.5-122B-A10B-OptiQ-2bit-REAP-63B

from mlx-community/Qwen3.5-122B-A10B-OptiQ-2bit
2-bit
Text generation ↓ 180 · ♥ 1
Params
65B
Weights
≈ 20.9 GB
Mac memory
32 GB+

apache-2.0 · updated 2026-09-15

Llama-3.2-3B-Instruct-abliterated-6bit

from huihui-ai/Llama-3.2-3B-Instruct-abliterated
6-bit
Text generation · huihui-ai ↓ 180 · ♥ 0
Params
3.2B
Weights
≈ 2.6 GB
Mac memory
8 GB+

llama3.2 · updated 2025-04-23

Qwen2.5-Math-7B-Instruct-4bit

from Qwen/Qwen2.5-Math-7B-Instruct
4-bit
Text generation · Qwen ↓ 180 · ♥ 0
Params
7.6B
Weights
≈ 4.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-09-18

Qwen3-Coder-30B-A3B-Instruct-8bit-DWQ-lr2e7

from Qwen/Qwen3-Coder-30B-A3B-Instruct
8-bit DWQ
Text generation · Qwen ↓ 180 · ♥ 0
Params
31B
Weights
≈ 32.4 GB
Mac memory
48 GB+

apache-2.0 · updated 2025-08-01

DeepSeek-Prover-V2-7B-4bit

from deepseek-ai/DeepSeek-Prover-V2-7B
4-bit
Text generation · DeepSeek ↓ 179 · ♥ 4
Params
6.9B
Weights
≈ 3.9 GB
Mac memory
8 GB+

updated 2025-05-01

Kimi-K2.5-mlx-DQ3_K_M-q8

from moonshotai/Kimi-K2.5
8-bit
Text generation · Moonshot · Kimi ↓ 179 · ♥ 3
Params
1.0T
Weights
≈ 1091 GB
Mac memory
> 512 GB

modified-mit · updated 2026-02-21

deepcogito-cogito-v1-preview-llama-3B-4bit

from deepcogito/cogito-v1-preview-llama-3B
4-bit
Text generation · deepcogito ↓ 179 · ♥ 0
Params
3.2B
Weights
≈ 1.8 GB
Mac memory
8 GB+

llama3.2 · updated 2025-04-08

gemma-2-9b-8bit

mlx-community/gemma-2-9b-8bit
8-bit
Text generation ↓ 177 · ♥ 9
Params
9.2B
Weights
≈ 9.8 GB
Mac memory
16 GB+

gemma · updated 2024-06-27

Qwen3-4B-4bit-DWQ

from Qwen/Qwen3-4B
4-bit DWQ
Text generation · Qwen ↓ 177 · ♥ 4
Params
4.0B
Weights
≈ 2.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-05-31

Qwen2.5-0.5B-Instruct-8bit

from Qwen/Qwen2.5-0.5B
8-bit
Text generation · Qwen ↓ 177 · ♥ 0
Params
494M
Weights
≈ 525 MB
Mac memory
8 GB+

apache-2.0 · updated 2024-09-18

Qwen2.5-Math-1.5B-Instruct-4bit

from Qwen/Qwen2.5-Math-1.5B-Instruct
4-bit
Text generation · Qwen ↓ 176 · ♥ 0
Params
1.5B
Weights
≈ 869 MB
Mac memory
8 GB+

apache-2.0 · updated 2024-09-18

Llama-3-8B-Instruct-1048k-8bit

mlx-community/Llama-3-8B-Instruct-1048k-8bit
8-bit
Text generation ↓ 175 · ♥ 17
Params
8.0B
Weights
≈ 8.5 GB
Mac memory
16 GB+

llama3 · updated 2024-04-29

MiniMax-M2-4bit

from MiniMaxAI/MiniMax-M2
4-bit
Text generation · MiniMaxAI ↓ 175 · ♥ 10
Params
229B
Weights
≈ 129 GB
Mac memory
192 GB+

mit · updated 2025-10-28

GLM-5.2-fp8

mlx-community/GLM-5.2-fp8
8-bit
Text generation ↓ 175 · ♥ 0
Params
743B
Weights
≈ 790 GB
Mac memory
> 512 GB

updated 2026-06-22

medgemma-27b-text-it-bf16

from google/medgemma-27b-text-it
bf16
Text generation · Google · Gemma ↓ 173 · ♥ 4
Params
28B
Weights
≈ 56.8 GB
Mac memory
96 GB+

health-ai-developer-foundations · updated 2025-06-09

Laguna-S-2.1-oQ4e-fast

from poolside/Laguna-S-2.1
4-bit
Text generation · poolside ↓ 173 · ♥ 3
Params
118B
Weights
≈ 66.2 GB
Mac memory
96 GB+

openmdw-1.1 · updated 2026-07-24

NVIDIA-Nemotron-3-Nano-30B-A3B-MLX-4Bit

from nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16
4-bit
Text generation · NVIDIA ↓ 173 · ♥ 1
Params
4.9B
Weights
≈ 4.2 GB
Mac memory
8 GB+

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

Devstral-Small-2507-4bit-DWQ

from mlx-community/Devstral-Samll-2507-bf16
4-bit DWQ
Text generation ↓ 172 · ♥ 12
Params
24B
Weights
≈ 13.3 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-07-12

Llama-4-Maverick-17B-16E-Instruct-4bit

from meta-llama/Llama-4-Maverick-17B-128E-Instruct
4-bit
Text generation · Meta · Llama ↓ 172 · ♥ 7
Params
401B
Weights
≈ 225 GB
Mac memory
256 GB+

llama4 · updated 2025-04-06

Qwen2.5.1-Coder-7B-Instruct-8bit

from Qwen/Qwen2.5-Coder-7B-Instruct
8-bit
Text generation · Qwen ↓ 172 · ♥ 3
Params
7.6B
Weights
≈ 8.1 GB
Mac memory
16 GB+

apache-2.0 · updated 2024-11-09

Llama-3.3-70B-Instruct-bf16

from meta-llama/Llama-3.3-70B-Instruct
bf16
Text generation · Meta · Llama ↓ 172 · ♥ 2
Params
71B
Weights
≈ 141 GB
Mac memory
192 GB+

llama3.3 · updated 2024-12-06

Qwen3-4B-3bit

from Qwen/Qwen3-4B
3-bit
Text generation · Qwen ↓ 172 · ♥ 0
Params
4.0B
Weights
≈ 1.8 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-04-28

Qwen2.5-Coder-14B-bf16

from Qwen/Qwen2.5-Coder-14B
bf16
Text generation · Qwen ↓ 171 · ♥ 1
Params
15B
Weights
≈ 29.5 GB
Mac memory
48 GB+

apache-2.0 · updated 2024-11-12

Qwen2.5-Coder-14B-8bit

from Qwen/Qwen2.5-Coder-14B
8-bit
Text generation · Qwen ↓ 171 · ♥ 0
Params
15B
Weights
≈ 15.7 GB
Mac memory
24 GB+

apache-2.0 · updated 2024-11-12

Qwen3-4B-Thinking-2507-heretic-4bit

from Daizee/Qwen3-4B-Thinking-2507-heretic
4-bit
Text generation · Daizee ↓ 170 · ♥ 0
Params
4.0B
Weights
≈ 2.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-02-16

DeepSeek-V3.2-8bit

from deepseek-ai/DeepSeek-V3.2
8-bit
Text generation · DeepSeek ↓ 169 · ♥ 0
Params
672B
Weights
≈ 714 GB
Mac memory
> 512 GB

mit · updated 2025-12-03

Llama-3.1-Nemotron-Nano-4B-v1.1-bf16

from nvidia/Llama-3.1-Nemotron-Nano-4B-v1.1
bf16
Text generation · NVIDIA ↓ 169 · ♥ 0
Params
4.5B
Weights
≈ 9.0 GB
Mac memory
16 GB+

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

Qwen3-4B-Instruct-2507-mxfp8

from Qwen/Qwen3-4B-Instruct-2507
MXFP8
Text generation · Qwen ↓ 168 · ♥ 0
Params
4.0B
Weights
≈ 4.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-02-08

Ling-mini-2.0-4bit

from inclusionAI/Ling-mini-2.0
4-bit
Text generation · inclusionAI ↓ 167 · ♥ 2
Params
16B
Weights
≈ 9.1 GB
Mac memory
16 GB+

mit · updated 2025-09-21

Qwen2.5-14B-Instruct-1M-abliterated-4bit

from huihui-ai/Qwen2.5-14B-Instruct-1M-abliterated
4-bit
Text generation · huihui-ai ↓ 167 · ♥ 1
Params
15B
Weights
≈ 8.3 GB
Mac memory
16 GB+

apache-2.0 · updated 2025-02-19

gpt-oss-20b-Ja-Fin-Thinking-MLX-4bit

from nri-ai/gpt-oss-20b-Ja-Fin-Thinking
4-bit
Text generation · nri-ai ↓ 167 · ♥ 1
Params
21B
Weights
≈ 11.8 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-03-27

gemma-2-2b-it-fp16

mlx-community/gemma-2-2b-it-fp16
fp16
Text generation ↓ 167 · ♥ 0
Params
2.6B
Weights
≈ 5.2 GB
Mac memory
8 GB+

gemma · updated 2024-11-06

Qwen2.5-7B-Instruct-bf16

from Qwen/Qwen2.5-7B
bf16
Text generation · Qwen ↓ 166 · ♥ 0
Params
7.6B
Weights
≈ 15.2 GB
Mac memory
24 GB+

apache-2.0 · updated 2024-09-18

SmolLM-135M-4bit

mlx-community/SmolLM-135M-4bit
4-bit
Text generation ↓ 165 · ♥ 2
Params
135M
Weights
≈ 76 MB
Mac memory
8 GB+

apache-2.0 · updated 2024-07-16

Meta-Llama-3.1-70B-Instruct-bf16-CORRECTED

from meta-llama/Llama-3.1-70B-Instruct
bf16
Text generation · Meta · Llama ↓ 165 · ♥ 0
Params
71B
Weights
≈ 141 GB
Mac memory
192 GB+

llama3.1 · updated 2024-10-05

bitnet-b1.58-2B-4T-4bit

from microsoft/bitnet-b1.58-2B-4T
4-bit
Text generation · Microsoft · Phi ↓ 165 · ♥ 0
Params
1.1B
Weights
≈ 631 MB
Mac memory
8 GB+

mit · updated 2025-06-10

Jinx-gpt-oss-20b-mxfp4-mlx

from Jinx-org/Jinx-gpt-oss-20b-mxfp4
MXFP4
Text generation · Jinx-org ↓ 164 · ♥ 4
Params
21B
Weights
≈ 14.4 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-09-17

Mistral-7B-v0.1-LoRA-Text2SQL

mlx-community/Mistral-7B-v0.1-LoRA-Text2SQL
Text generation ↓ 164 · ♥ 3
Params
—
Weights
—
Mac memory
—

mit · updated 2024-01-29