Skip to content
AI Model Radar
Apple silicon

Apple MLX models

5,842 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

5,842 builds · page 69 of 98

gemma-4-E4B-it-qat-mxfp8

from google/gemma-4-e4b-it-qat-q4_0-unquantized
MXFP8
Any to any · Google · Gemma ↓ 33 · ♥ 0
Params
7.9B
Weights
≈ 8.9 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-06-05

granite-4.0-1b-3bit

from ibm-granite/granite-4.0-1b
3-bit
Text generation · IBM · Granite ↓ 33 · ♥ 0
Params
1.6B
Weights
≈ 714 MB
Mac memory
8 GB+

apache-2.0 · updated 2025-10-28

granite-4.0-1b-6bit

from ibm-granite/granite-4.0-1b
6-bit
Text generation · IBM · Granite ↓ 33 · ♥ 0
Params
1.6B
Weights
≈ 1.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-10-28

granite-4.0-1b-bf16

from ibm-granite/granite-4.0-1b
bf16
Text generation · IBM · Granite ↓ 33 · ♥ 0
Params
1.6B
Weights
≈ 3.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-10-28

nanoLLaVA-1.5-6bit

mlx-community/nanoLLaVA-1.5-6bit
6-bit
Uncategorised ↓ 33 · ♥ 0
Params
1.1B
Weights
≈ 857 MB
Mac memory
8 GB+

apache-2.0 · updated 2024-12-26

nanoLLaVA-1.5-bf16

mlx-community/nanoLLaVA-1.5-bf16
bf16
Uncategorised ↓ 33 · ♥ 0
Params
1.1B
Weights
≈ 2.1 GB
Mac memory
8 GB+

apache-2.0 · updated 2024-12-26

omost-phi-3-mini-128k-mlx-8bit

mlx-community/omost-phi-3-mini-128k-mlx-8bit
8-bit
Uncategorised ↓ 33 · ♥ 0
Params
3.8B
Weights
≈ 4.1 GB
Mac memory
8 GB+

updated 2024-06-07

sarashina2.2-0.5b-instruct-v0.1

from sbintuitions/sarashina2.2-0.5b-instruct-v0.1
Text generation · sbintuitions ↓ 33 · ♥ 0
Params
793M
Weights
≈ 1.6 GB
Mac memory
8 GB+

mit · updated 2025-07-03

stockmark-100b-4bit

mlx-community/stockmark-100b-4bit
4-bit
Text generation ↓ 33 · ♥ 0
Params
96B
Weights
≈ 54.1 GB
Mac memory
64 GB+

mit · updated 2024-05-16

whisper-base.en-4bit

from openai/whisper-base.en
4-bit
Speech recognition · OpenAI · gpt-oss ↓ 33 · ♥ 0
Params
72M
Weights
≈ 42 MB
Mac memory
8 GB+

apache-2.0 · updated 2025-12-15

whisper-base.en-mlx-8bit

mlx-community/whisper-base.en-mlx-8bit
8-bit
Uncategorised ↓ 33 · ♥ 0
Params
—
Weights
—
Mac memory
—

updated 2024-03-09

GLM-4.5V-8bit

mlx-community/GLM-4.5V-8bit
8-bit
Uncategorised ↓ 32 · ♥ 2
Params
—
Weights
—
Mac memory
—

mit · updated 2025-08-11

ShowUI-2B-bf16

from Qwen/Qwen2-VL-2B-Instruct
bf16
Uncategorised · Qwen ↓ 32 · ♥ 2
Params
2.2B
Weights
≈ 4.4 GB
Mac memory
8 GB+

mit · updated 2025-02-26

Starling-LM-7B-beta

mlx-community/Starling-LM-7B-beta
Text generation ↓ 32 · ♥ 2
Params
7.2B
Weights
≈ 28.7 GB
Mac memory
48 GB+

apache-2.0 · updated 2024-03-28

dolphin-2.9-llama3-8b-8bit

from meta-llama/Meta-Llama-3-8B
8-bit
Uncategorised · Meta · Llama ↓ 32 · ♥ 2
Params
8.0B
Weights
≈ 8.5 GB
Mac memory
16 GB+

other · updated 2024-04-21

sam3-mxfp8

from facebook/sam3
MXFP8
Uncategorised · facebook ↓ 32 · ♥ 2
Params
860M
Weights
≈ 985 MB
Mac memory
8 GB+

updated 2026-03-28

FlashVSR-v1.1-fp32

from JunhaoZhuang/FlashVSR-v1.1
fp32
Video to video · JunhaoZhuang ↓ 32 · ♥ 1
Params
—
Weights
—
Mac memory
—

apache-2.0 · updated 2026-10-05

INTELLECT-1-Instruct-6bit

from PrimeIntellect/INTELLECT-1-Instruct
6-bit
Text generation · PrimeIntellect ↓ 32 · ♥ 1
Params
10B
Weights
≈ 8.3 GB
Mac memory
16 GB+

apache-2.0 · updated 2024-12-02

LFM2-700M-5bit

from LiquidAI/LFM2-700M
5-bit
Text generation · LiquidAI ↓ 32 · ♥ 1
Params
742M
Weights
≈ 511 MB
Mac memory
8 GB+

lfm1.0 · updated 2025-07-11

Ling-2.6-flash-mlx-5bit

from inclusionAI/Ling-2.6-flash
5-bit
Text generation · inclusionAI ↓ 32 · ♥ 1
Params
104B
Weights
≈ 71.6 GB
Mac memory
96 GB+

mit · updated 2026-04-30

MIST-1-70B-MLX

from NousResearch/Hermes-3-Llama-3.1-70B
Text generation · NousResearch ↓ 32 · ♥ 1
Params
71B
Weights
≈ 141 GB
Mac memory
192 GB+

llama3.1 · updated 2026-05-31

OlympicCoder-32B-4bit

from open-r1/OlympicCoder-32B
4-bit
Text generation · open-r1 ↓ 32 · ♥ 1
Params
33B
Weights
≈ 18.4 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-03-19

Phi-4-reasoning-6bit

from microsoft/Phi-4-reasoning
6-bit
Text generation · Microsoft · Phi ↓ 32 · ♥ 1
Params
15B
Weights
≈ 11.9 GB
Mac memory
16 GB+

mit · updated 2025-05-01

Pleias-RAG-1B-bf16

from PleIAs/Pleias-RAG-1B
bf16
Text generation · PleIAs ↓ 32 · ♥ 1
Params
1.2B
Weights
≈ 2.4 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-04-24

Polaris-4B-Preview-4bit

from POLARIS-Project/Polaris-4B-Preview
4-bit
Text generation · POLARIS-Project ↓ 32 · ♥ 1
Params
4.0B
Weights
≈ 2.3 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-07-03

Qwopus3.6-27B-Coder-6bit

from Jackrong/Qwopus3.6-27B-Coder
6-bit
Vision + text · Jackrong ↓ 32 · ♥ 1
Params
27B
Weights
≈ 22.8 GB
Mac memory
32 GB+

apache-2.0 · updated 2026-07-01

ether0-8bit

from futurehouse/ether0
8-bit
Text generation · futurehouse ↓ 32 · ♥ 1
Params
24B
Weights
≈ 25.0 GB
Mac memory
32 GB+

apache-2.0 · updated 2025-06-05

gemma-3-12b-it-qat-8bit

from OpenGVLab/InternVL3-1B-Instruct
8-bit
Vision + text · OpenGVLab ↓ 32 · ♥ 1
Params
—
Weights
—
Mac memory
—

qwen · updated 2025-04-21

gemma-3-27b-pt-8bit

mlx-community/gemma-3-27b-pt-8bit
8-bit
Vision + text ↓ 32 · ♥ 1
Params
—
Weights
—
Mac memory
—

gemma · updated 2025-03-18

gemma-3-4b-it-qat-bf16

from google/gemma-3-4b-it
bf16
Vision + text · Google · Gemma ↓ 32 · ♥ 1
Params
5.0B
Weights
≈ 9.9 GB
Mac memory
16 GB+

gemma · updated 2025-04-18

gemma-4-e2b-nvfp4

from google/gemma-4-e2b
4-bit
Any to any · Google · Gemma ↓ 32 · ♥ 1
Params
5.1B
Weights
≈ 3.6 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-04-03

granite-3.3-2b-instruct-fp16

from ibm-granite/granite-3.3-2b-instruct
fp16
Text generation · IBM · Granite ↓ 32 · ♥ 1
Params
2.5B
Weights
≈ 5.1 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-04-16

Big-Alice-28B-v1-bf16

from TheDrummer/Big-Alice-28B-v1
bf16
Text generation · TheDrummer ↓ 32 · ♥ 0
Params
29B
Weights
≈ 57.2 GB
Mac memory
96 GB+

mit · updated 2025-06-08

CAT-Thinking-8B-MLX-6bit

from cyberagent/CAT-Thinking-8B
6-bit
Text generation · cyberagent ↓ 32 · ♥ 0
Params
8.2B
Weights
≈ 6.7 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-06-03

CausalLM-35b-beta-long-4bit

from CausalLM/35b-beta-long
4-bit
Uncategorised · CausalLM ↓ 32 · ♥ 0
Params
35B
Weights
≈ 19.7 GB
Mac memory
24 GB+

wtfpl · updated 2024-12-14

DeepHermes-3-Llama-3-3B-Preview-bf16

from NousResearch/DeepHermes-3-Llama-3-3B-Preview
bf16
Text generation · NousResearch ↓ 32 · ♥ 0
Params
3.2B
Weights
≈ 6.4 GB
Mac memory
16 GB+

llama3 · updated 2025-03-14

EVA-Qwen2.5-72B-v0.0-8bit

from EVA-UNIT-01/EVA-Qwen2.5-72B-v0.0
8-bit
Text generation · EVA-UNIT-01 ↓ 32 · ♥ 0
Params
73B
Weights
≈ 77.3 GB
Mac memory
96 GB+

qwen · updated 2024-10-31

Falcon-H1-1.5B-Instruct-8bit

from tiiuae/Falcon-H1-1.5B-Instruct
8-bit
Text generation · TII · Falcon ↓ 32 · ♥ 0
Params
1.6B
Weights
≈ 1.7 GB
Mac memory
8 GB+

falcon-llm-license · updated 2025-09-26

Falcon-H1-1.5B-Instruct-bf16

from tiiuae/Falcon-H1-1.5B-Instruct
bf16
Text generation · TII · Falcon ↓ 32 · ♥ 0
Params
1.6B
Weights
≈ 3.1 GB
Mac memory
8 GB+

falcon-llm-license · updated 2025-06-16

Falcon3-7B-Instruct-8bit

from tiiuae/Falcon3-7B-Instruct
8-bit
Text generation · TII · Falcon ↓ 32 · ♥ 0
Params
7.5B
Weights
≈ 7.9 GB
Mac memory
16 GB+

falcon-llm-license · updated 2024-12-18

Gelato-30B-A3B-4bit

from mlfoundations/Gelato-30B-A3B
4-bit
Vision + text · mlfoundations ↓ 32 · ♥ 0
Params
—
Weights
—
Mac memory
—

apache-2.0 · updated 2025-11-17

IQuest-Coder-V1-40B-Instruct-5bit

from IQuestLab/IQuest-Coder-V1-40B-Instruct
5-bit
Text generation · IQuestLab ↓ 32 · ♥ 0
Params
40B
Weights
≈ 27.4 GB
Mac memory
48 GB+

iquestcoder · updated 2026-01-02

Josiefied-Qwen3.5-0.8B-gabliterated-v1-6bit

from Goekdeniz-Guelmez/Josiefied-Qwen3.5-0.8B-gabliterated-v1
6-bit
Text generation · Goekdeniz-Guelmez ↓ 32 · ♥ 0
Params
752M
Weights
≈ 612 MB
Mac memory
8 GB+

updated 2026-03-13

Klear-46B-A2.5B-Instruct-3bit

from Kwai-Klear/Klear-46B-A2.5B-Instruct
3-bit
Text generation · Kwai-Klear ↓ 32 · ♥ 0
Params
46B
Weights
≈ 20.2 GB
Mac memory
32 GB+

apache-2.0 · updated 2025-09-08

LFM2-2.6B-Transcript-5bit

from LiquidAI/LFM2-2.6B-Transcript
5-bit
Text generation · LiquidAI ↓ 32 · ♥ 0
Params
2.6B
Weights
≈ 1.8 GB
Mac memory
8 GB+

lfm1.0 · updated 2026-01-07

Llama-3-16B-Instruct-v0.1-4bit

mlx-community/Llama-3-16B-Instruct-v0.1-4bit
4-bit
Text generation ↓ 32 · ♥ 0
Params
3.1B
Weights
≈ 3.2 GB
Mac memory
8 GB+

updated 2024-04-20

Llama-3-8b-64k-PoSE-4bit

mlx-community/Llama-3-8b-64k-PoSE-4bit
4-bit
Text generation ↓ 32 · ♥ 0
Params
8.0B
Weights
≈ 4.5 GB
Mac memory
8 GB+

updated 2024-04-25

MAI-UI-8B-6bit

from Tongyi-MAI/MAI-UI-8B
6-bit
Vision + text · Tongyi-MAI ↓ 32 · ♥ 0
Params
8.8B
Weights
≈ 9.0 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-03-20

MiniCPM3-4B-6bit

from Goekdeniz-Guelmez/OpenBNB-MiniCPM3-4b
6-bit
Text generation · Goekdeniz-Guelmez ↓ 32 · ♥ 0
Params
4.3B
Weights
≈ 3.5 GB
Mac memory
8 GB+

apache-2.0 · updated 2025-03-19

MiniCPM3-4B-bfloat16

from Goekdeniz-Guelmez/OpenBNB-MiniCPM3-4b
Text generation · Goekdeniz-Guelmez ↓ 32 · ♥ 0
Params
4.3B
Weights
≈ 8.5 GB
Mac memory
16 GB+

apache-2.0 · updated 2025-03-19

Nova-Tempus-70B-v0.1-MLX-3Bit

from sophosympatheia/Nova-Tempus-70B-v0.1
3-bit
Text generation · sophosympatheia ↓ 32 · ♥ 0
Params
71B
Weights
≈ 30.9 GB
Mac memory
48 GB+

llama3.3 · updated 2025-01-26

OpenReasoning-Nemotron-32B-4bit

from nvidia/OpenReasoning-Nemotron-32B
4-bit
Text generation · NVIDIA ↓ 32 · ♥ 0
Params
33B
Weights
≈ 65.5 GB
Mac memory
96 GB+

cc-by-4.0 · updated 2025-07-22

Preferred-MedLLM-Qwen-72B-8bit

from pfnet/Preferred-MedLLM-Qwen-72B
8-bit
Text generation · pfnet ↓ 32 · ♥ 0
Params
73B
Weights
≈ 77.3 GB
Mac memory
96 GB+

qwen · updated 2025-03-08

Qwen-3-VL-Ricoh-8B-20260227-MLX-8bit

from ricoh-ai/Qwen-3-VL-Ricoh-8B-20260227
8-bit
Vision + text · ricoh-ai ↓ 32 · ♥ 0
Params
8.8B
Weights
≈ 9.9 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-03-30

Qwen1.5-72B-4bit

mlx-community/Qwen1.5-72B-4bit
4-bit
Text generation ↓ 32 · ♥ 0
Params
72B
Weights
≈ 42.5 GB
Mac memory
64 GB+

tongyi-qianwen · updated 2024-04-13

Qwen3-4B-Sky-High-Hermes-bfloat16

from ZeroXClem/Qwen3-4B-Sky-High-Hermes
Text generation · ZeroXClem ↓ 32 · ♥ 0
Params
4.0B
Weights
≈ 8.0 GB
Mac memory
16 GB+

apache-2.0 · updated 2026-01-19

Qwen3-TTS-12Hz-1.7B-CustomVoice-5bit

mlx-community/Qwen3-TTS-12Hz-1.7B-CustomVoice-5bit
5-bit
Text to speech ↓ 32 · ♥ 0
Params
1.9B
Weights
≈ 1.8 GB
Mac memory
8 GB+

apache-2.0 · updated 2026-01-25

Qwen3-VL-8B-Thinking-bf16

mlx-community/Qwen3-VL-8B-Thinking-bf16
bf16
Vision + text ↓ 32 · ♥ 0
Params
8.8B
Weights
≈ 17.5 GB
Mac memory
24 GB+

apache-2.0 · updated 2025-10-14