Apple MLX models

DeepSeek-R1-Distill-Qwen-1.5B-bf16

from deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B

bf16

Uncategorised · DeepSeek ↓ 71 · ♥ 1

Params

1.8B

Weights

≈ 3.6 GB

Mac memory

8 GB+

updated 2025-02-26

Hermes-2-Pro-Llama-3-8B-4bit

from NousResearch/Meta-Llama-3-8B

4-bit

Uncategorised · NousResearch ↓ 71 · ♥ 1

Params

8.0B

Weights

≈ 4.5 GB

Mac memory

8 GB+

apache-2.0 · updated 2024-05-02

Qwen2-VL-7B-Instruct-abliterated-8bit

from Qwen/Qwen2-VL-7B-Instruct

8-bit

Vision + text · Qwen ↓ 71 · ♥ 1

Params

8.3B

Weights

≈ 8.8 GB

Mac memory

16 GB+

apache-2.0 · updated 2025-03-24

Qwen3-VL-Embedding-2B-mxfp8

from Qwen/Qwen3-VL-2B-Instruct

MXFP8

Embeddings · Qwen ↓ 71 · ♥ 1

Params

2.1B

Weights

≈ 2.6 GB

Mac memory

8 GB+

apache-2.0 · updated 2026-03-14

codegemma-1.1-2b-4bit

from google/codegemma-1.1-2b

4-bit

Text generation · Google · Gemma ↓ 71 · ♥ 1

Params

2.5B

Weights

≈ 1.4 GB

Mac memory

8 GB+

gemma · updated 2025-06-02

gemma-4-12B-5bit

from google/gemma-4-12B

5-bit

Any to any · Google · Gemma ↓ 71 · ♥ 1

Params

12B

Weights

≈ 8.2 GB

Mac memory

16 GB+

apache-2.0 · updated 2026-06-03

llm-jp-3-8x1.8b-instruct3-4bit

from llm-jp/llm-jp-3-8x1.8b-instruct3

4-bit

Text generation · llm-jp ↓ 71 · ♥ 1

Params

9.3B

Weights

≈ 5.2 GB

Mac memory

8 GB+

apache-2.0 · updated 2025-03-27

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

from llm-jp/llm-jp-4-8b-thinking

8-bit

Text generation · llm-jp ↓ 71 · ♥ 1

Params

8.6B

Weights

≈ 9.1 GB

Mac memory

16 GB+

apache-2.0 · updated 2026-04-12

1.5-Pints-2K-v0.1

from pints-ai/1.5-Pints-2K-v0.1

Text generation · pints-ai ↓ 71 · ♥ 0

Params

1.6B

Weights

≈ 3.1 GB

Mac memory

8 GB+

mit · updated 2024-10-28

3b-es_it-ft-research_release-8bit

from canopylabs/orpheus-3b-0.1-pretrained

8-bit

Text to speech · canopylabs ↓ 71 · ♥ 0

Params

3.3B

Weights

≈ 3.5 GB

Mac memory

8 GB+

apache-2.0 · updated 2025-04-09

AutoGLM-Phone-9B-Multilingual-4bit

from zai-org/GLM-4.1V-9B-Base

4-bit

Vision + text · zai-org ↓ 71 · ♥ 0

Params

—

Weights

—

Mac memory

—

mit · updated 2025-12-12

Dobby-Unhinged-Llama-3.3-70B-mlx-4Bit

from SentientAGI/Dobby-Unhinged-Llama-3.3-70B

4-bit

Text generation · SentientAGI ↓ 71 · ♥ 0

Params

71B

Weights

≈ 39.7 GB

Mac memory

48 GB+

llama3.3 · updated 2025-09-15

GLM-4.6V-6bit

from zai-org/GLM-4.6V

6-bit

Vision + text · zai-org ↓ 71 · ♥ 0

Params

—

Weights

—

Mac memory

—

mit · updated 2026-02-22

Hermes-3-Llama-3.1-8B-3bit

from NousResearch/Hermes-3-Llama-3.1-8B

3-bit

Uncategorised · NousResearch ↓ 71 · ♥ 0

Params

8.0B

Weights

≈ 3.5 GB

Mac memory

8 GB+

llama3 · updated 2024-12-13

Llama-3.1-Tulu-3-70B-4bit

from allenai/Llama-3.1-Tulu-3-70B

4-bit

Text generation · Ai2 · OLMo ↓ 71 · ♥ 0

Params

71B

Weights

≈ 39.7 GB

Mac memory

48 GB+

llama3.1 · updated 2024-11-25

Mellum2-12B-A2.5B-Thinking-8bit

from JetBrains/Mellum2-12B-A2.5B-Thinking

8-bit

Text generation · JetBrains ↓ 71 · ♥ 0

Params

12B

Weights

≈ 12.9 GB

Mac memory

24 GB+

apache-2.0 · updated 2026-08-11

Ministral-3-8B-Reasoning-2512-8bit

from mistralai/Ministral-3-8B-Base-2512

8-bit

Uncategorised · Mistral ↓ 71 · ♥ 0

Params

—

Weights

—

Mac memory

—

apache-2.0 · updated 2025-12-03

Qwen2.5-7B-Instruct-1M-abliterated-8bit

from huihui-ai/Qwen2.5-7B-Instruct-1M-abliterated

8-bit

Text generation · huihui-ai ↓ 71 · ♥ 0

Params

7.6B

Weights

≈ 8.1 GB

Mac memory

16 GB+

apache-2.0 · updated 2025-02-19

Qwen3-4B-Instruct-2507-5bit-g32

from Qwen/Qwen3-4B-Instruct-2507

5-bit

Text generation · Qwen ↓ 71 · ♥ 0

Params

4.0B

Weights

≈ 2.8 GB

Mac memory

8 GB+

apache-2.0 · updated 2025-10-11

Solar-Open-100B-4bit

from upstage/Solar-Open-100B

4-bit

Text generation · upstage ↓ 71 · ♥ 0

Params

103B

Weights

≈ 57.8 GB

Mac memory

96 GB+

solar-apache-2.0 · updated 2026-01-06

c4ai-command-r-plus-08-2024-4bit

mlx-community/c4ai-command-r-plus-08-2024-4bit

4-bit

Text generation ↓ 71 · ♥ 0

Params

104B

Weights

≈ 58.4 GB

Mac memory

96 GB+

cc-by-nc-4.0 · updated 2024-08-31

granite-4.0-micro-8bit

from ibm-granite/granite-4.0-micro

8-bit

Text generation · IBM · Granite ↓ 71 · ♥ 0

Params

3.4B

Weights

≈ 3.6 GB

Mac memory

8 GB+

apache-2.0 · updated 2025-10-02

llm-jp-13b-instruct-full-ac_001-dolly-ichikara_004_001_single-oasst-oasst2-v2.0

mlx-community/llm-jp-13b-instruct-full-ac_001-dolly-ichikara_004_001_single-oasst-oasst2-v2.0

Text generation ↓ 71 · ♥ 0

Params

14B

Weights

≈ 27.4 GB

Mac memory

48 GB+

apache-2.0 · updated 2024-04-30

Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-4.5bit-msq

mlx-community/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-4.5bit-msq

5-bit

Vision + text ↓ 70 · ♥ 11

Params

40B

Weights

≈ 6.5 GB

Mac memory

16 GB+

apache-2.0 · updated 2026-05-01

MiniMax-M2-8bit

from MiniMaxAI/MiniMax-M2

8-bit

Text generation · MiniMaxAI ↓ 70 · ♥ 8

Params

229B

Weights

≈ 243 GB

Mac memory

512 GB+

mit · updated 2025-10-28

Bernini-R-int4

from ByteDance/Bernini-R-Diffusers

Text to video · ByteDance ↓ 70 · ♥ 6

Params

—

Weights

—

Mac memory

—

apache-2.0 · updated 2026-06-04

Ouro-2.6B-Thinking-4bit

from ByteDance/Ouro-2.6B-Thinking

4-bit

Text generation · ByteDance ↓ 70 · ♥ 4

Params

2.7B

Weights

≈ 1.5 GB

Mac memory

8 GB+

apache-2.0 · updated 2025-11-28

granite-4.0-1b-speech-4bit

from ibm-granite/granite-4.0-1b-base

4-bit

Uncategorised · IBM · Granite ↓ 70 · ♥ 4

Params

2.3B

Weights

≈ 2.0 GB

Mac memory

8 GB+

apache-2.0 · updated 2026-03-11

gemma-3n-E2B-bf16

from google/gemma-3n-E4B

bf16

Vision + text · Google · Gemma ↓ 70 · ♥ 3

Params

6.0B

Weights

≈ 10.9 GB

Mac memory

16 GB+

gemma · updated 2025-07-12

LongWriter-llama3.1-8b-6bit

from zai-org/LongWriter-llama3.1-8b

6-bit

Text generation · zai-org ↓ 70 · ♥ 2

Params

8.0B

Weights

≈ 6.5 GB

Mac memory

16 GB+

llama3.1 · updated 2025-09-28

granite-4.0-h-micro-8bit

from ibm-granite/granite-4.0-h-micro

8-bit

Text generation · IBM · Granite ↓ 70 · ♥ 2

Params

3.2B

Weights

≈ 3.4 GB

Mac memory

8 GB+

apache-2.0 · updated 2025-10-02

Apertus-8B-Instruct-2509-8bit

from swiss-ai/Apertus-8B-Instruct-2509

8-bit

Text generation · swiss-ai ↓ 70 · ♥ 1

Params

8.1B

Weights

≈ 8.6 GB

Mac memory

16 GB+

apache-2.0 · updated 2025-09-03

NeuralDaredevil-8B-abliterated-4bit

mlx-community/NeuralDaredevil-8B-abliterated-4bit

4-bit

Uncategorised ↓ 70 · ♥ 1

Params

8.0B

Weights

≈ 4.5 GB

Mac memory

8 GB+

other · updated 2024-07-13

Qwen3-30B-A3B-Instruct-2507-4bit-ODWQ

from Qwen/Qwen3-30B-A3B

4-bit

Text generation · Qwen ↓ 70 · ♥ 1

Params

31B

Weights

≈ 17.2 GB

Mac memory

24 GB+

apache-2.0 · updated 2025-08-16

3b-es_it-ft-research_release-bf16

from canopylabs/orpheus-3b-0.1-pretrained

bf16

Text to speech · canopylabs ↓ 70 · ♥ 0

Params

—

Weights

—

Mac memory

—

apache-2.0 · updated 2025-04-09

Dia2-2B-bf16

from nari-labs/Dia2-2B

bf16

Text to speech · nari-labs ↓ 70 · ♥ 0

Params

1.9B

Weights

≈ 3.8 GB

Mac memory

8 GB+

apache-2.0 · updated 2026-10-07

Florence-2-large-ft-4bit

mlx-community/Florence-2-large-ft-4bit

4-bit

Vision + text ↓ 70 · ♥ 0

Params

823M

Weights

≈ 504 MB

Mac memory

8 GB+

mit · updated 2024-11-21

Ministral-3-3B-Instruct-2512-6bit

from mistralai/Ministral-3-3B-Base-2512

6-bit

Uncategorised · Mistral ↓ 70 · ♥ 0

Params

3.8B

Weights

≈ 3.6 GB

Mac memory

8 GB+

apache-2.0 · updated 2025-12-03

Qwen2.5-0.5B-Instruct-2bit-mlx

from Qwen/Qwen2.5-0.5B-Instruct

2-bit

Text generation · Qwen ↓ 70 · ♥ 0

Params

494M

Weights

≈ 155 MB

Mac memory

8 GB+

apache-2.0 · updated 2025-02-17

Qwen2.5-VL-3B-Instruct-abliterated-6bit

from Qwen/Qwen2.5-VL-3B-Instruct

6-bit

Vision + text · Qwen ↓ 70 · ♥ 0

Params

3.8B

Weights

≈ 3.2 GB

Mac memory

8 GB+

updated 2025-02-25

all-MiniLM-L6-v2-8bit

mlx-community/all-MiniLM-L6-v2-8bit

8-bit

Sentence similarity ↓ 70 · ♥ 0

Params

23M

Weights

≈ 24 MB

Mac memory

8 GB+

apache-2.0 · updated 2025-04-02

deepseek-vl2-bf16

mlx-community/deepseek-vl2-bf16

bf16

Vision + text ↓ 70 · ♥ 0

Params

27B

Weights

≈ 55.0 GB

Mac memory

64 GB+

deepseek · updated 2024-12-22

kanana-2-3b-instruct-8bit

from kakaocorp/kanana-2-3b-instruct

8-bit

Text generation · kakaocorp ↓ 70 · ♥ 0

Params

3.2B

Weights

≈ 3.4 GB

Mac memory

8 GB+

kanana-open-license · updated 2026-08-06

granite-4.1-30b-mxfp8

from ibm-granite/granite-4.1-30b

MXFP8

Text generation · IBM · Granite ↓ 69 · ♥ 3

Params

29B

Weights

≈ 30.7 GB

Mac memory

48 GB+

apache-2.0 · updated 2026-04-30

DeepSeek-V3.1-Terminus-4bit

from deepseek-ai/DeepSeek-V3.1-Terminus

4-bit

Text generation · DeepSeek ↓ 69 · ♥ 2

Params

671B

Weights

≈ 378 GB

Mac memory

512 GB+

mit · updated 2025-09-22

Ornith-1.0-9B-8bit

from ornith-ai/Ornith-1.0-9B

8-bit

Text generation · ornith-ai ↓ 69 · ♥ 2

Params

9.4B

Weights

≈ 10.4 GB

Mac memory

16 GB+

mit · updated 2026-06-28

Qwen3.5-27B-5bit

mlx-community/Qwen3.5-27B-5bit

5-bit

Vision + text ↓ 69 · ♥ 2

Params

27B

Weights

≈ 19.4 GB

Mac memory

24 GB+

apache-2.0 · updated 2026-02-24

Stockmark-2-100B-Instruct-beta-4bit

from stockmark/Stockmark-2-100B-Instruct-beta

4-bit

Text generation · stockmark ↓ 69 · ♥ 2

Params

96B

Weights

≈ 54.0 GB

Mac memory

64 GB+

mit · updated 2025-03-08

echo-tts-base

from jordand/echo-tts-base

Uncategorised · jordand ↓ 69 · ♥ 2

Params

2.8B

Weights

≈ 5.6 GB

Mac memory

16 GB+

cc-by-nc-sa-4.0 · updated 2026-02-27

ABEJA-Qwen3-14B-Agentic-256k-v0.1-MLX-8bit

from abeja/ABEJA-Qwen3-14B-Agentic-256k-v0.1

8-bit

Text generation · abeja ↓ 69 · ♥ 1

Params

15B

Weights

≈ 15.7 GB

Mac memory

24 GB+

apache-2.0 · updated 2026-03-28

gemma-4-E4B-it-qat-mxfp4

from google/gemma-4-e4b-it-qat-q4_0-unquantized

MXFP4

Any to any · Google · Gemma ↓ 69 · ♥ 1

Params

7.9B

Weights

≈ 5.1 GB

Mac memory

8 GB+

apache-2.0 · updated 2026-06-05

sam-audio-base-fp16

from facebook/sam-audio-base

fp16

Audio to audio · facebook ↓ 69 · ♥ 1

Params

1.3B

Weights

≈ 2.5 GB

Mac memory

8 GB+

other · updated 2026-02-06

snowflake-arctic-embed-l-v2.0-bf16

mlx-community/snowflake-arctic-embed-l-v2.0-bf16

bf16

Sentence similarity ↓ 69 · ♥ 1

Params

568M

Weights

≈ 1.1 GB

Mac memory

8 GB+

apache-2.0 · updated 2025-04-02

3b-zh-ft-research_release-6bit

from canopylabs/orpheus-3b-0.1-pretrained

6-bit

Text to speech · canopylabs ↓ 69 · ♥ 0

Params

3.3B

Weights

≈ 2.7 GB

Mac memory

8 GB+

apache-2.0 · updated 2025-04-09

DeepSeek-R1-0528-3bit

from deepseek-ai/DeepSeek-R1-0528

3-bit

Text generation · DeepSeek ↓ 69 · ♥ 0

Params

671B

Weights

≈ 294 GB

Mac memory

512 GB+

mit · updated 2025-05-29

Falcon-H1-Tiny-Coder-90M-4bit

from tiiuae/Falcon-H1-Tiny-Coder-90M

4-bit

Text generation · TII · Falcon ↓ 69 · ♥ 0

Params

91M

Weights

≈ 51 MB

Mac memory

8 GB+

falcon-llm-license · updated 2026-01-28

GLM-4.6V-heretic-nvfp4

from coder3101/GLM-4.6V-heretic

4-bit

Vision + text · coder3101 ↓ 69 · ♥ 0

Params

108B

Weights

≈ 61.9 GB

Mac memory

96 GB+

updated 2026-03-28

Josiefied-Qwen3-0.6B-abliterated-v1-4bit

from Goekdeniz-Guelmez/Josiefied-Qwen3-0.6B-abliterated-v1

4-bit

Text generation · Goekdeniz-Guelmez ↓ 69 · ♥ 0

Params

596M

Weights

≈ 335 MB

Mac memory

8 GB+

updated 2025-04-29

Llama-4-Scout-17B-16E-8bit

mlx-community/Llama-4-Scout-17B-16E-8bit

8-bit

Vision + text ↓ 69 · ♥ 0

Params

—

Weights

—

Mac memory

—

llama4 · updated 2025-04-06

MedraN-E4B-Uncensored-Q4

from nicoboss/MedraN-E4B-Uncensored-EP7

4-bit

Uncategorised · nicoboss ↓ 69 · ♥ 0

Params

6.9B

Weights

≈ 3.9 GB

Mac memory

8 GB+

custom-license · updated 2026-02-15