Hugging Face's logo
mlx-community 's Collections
Qwen-Image-Flash (MLX)
Laguna-S-2.1
Ministral-3 Base
Step 3.7 Flash
Inkling MLX
Hy3
FLUX.2 klein (MLX)
Z-Image (MLX)
Laguna-XS-2.1
Agents-A1
Ornith 1.0
Inpainting (MLX)
DDColor (MLX)
BiRefNet (MLX)
VibeThinker-3B
Boogu-Image-0.1 (MLX)
Gemma-4-12b-coder-fable5-composer2.5
DiffusionGemma
SeedVR2 (MLX-Swift)
V-JEPA2 (MLX)
SigLIP2 NR-IQA MLX
Real-ESRGAN (MLX)
Cocktail-Fork MRX (MLX)
Gemma 4 MTP QAT
Gemma 4 QAT
EfRLFN MLX
NAFNet MLX
Bernini-R MLX
LongCat-Video — MLX
Abliterated Huihui LFM2.5-8B-A1B (MLX)
Qwen 3.x MTP
Lens 3.8B (MLX)
SoulX-Singer MLX
LongCat-Video-Avatar 1.5 — MLX
SongGeneration v2 MLX
Hy-MT2
Supertonic 3
Lance MLX
PRIME-RL P1-VL-30B-A3B
Simplified Reasoning SU-01
OmniVoice
MiMo-V2.5-ASR
MiniCPM-V 4.6
Gemma-4 Assistant (MTP)
Hy3 preview
Granite 4.1
UNCENSORED Qwen 3.6 27B
inclusionAI LING 2.6
inclusionAI DR-Venus
Nvidia Nemotron-3-Nano-Omni
Mel-Band-RoFormer (MLX)
Poolside Laguna-XS.2
DeepSeek V4
MiniMax-M2.7
YOLO 26
Glm 5.1
Gemma 4
Nemotron-Cascade 2
Demucs MLX — Music Source Separation
KittenTTS
Qwen-3.5
Step 3.5 Flash
MedGemma-1.5
PaddleOCR-VL
Qwen3-ASR
Qwen3-TTS
facebook Perception LM
IQuest-Coder
K-EXAONE
Solar Open
Gabliterated v1
Sam Audio
Perception Encoder
FunctionGemma
Chatterbox TTS
GPT-OSS
VoxCPM
VibeVoice
GLM-4.6V
Orchestrator 8B
INTELLECT 3
Josiefied and Abliterated Models
Olmo-3
Granite 4.0 Nano Language Models
LLaDA 2.0
MiniMax-M2
olmOCR 2
Qwen3-VL
Nanonets OCR2
💧LFM2-8B-A1B-MoE
ServiceNow-Apriel
Qwen3-Coder-MoE
Granite-4.0 Family
GLM-4.6
GLM-4.5-Air
Qwen3 Next
SEA-LION
Lille 130M
EmbeddingGemma
Apertus
Swahili Gemma 1B
LFM2-VL
Gemma 3-270m
VisualQuality-R1
GLM 4.5
olmOCR-0725
LFM2.x
SmolLM3
ERNIE-4.5
DiffuCoder-7B
JetBrains Mellum
Gemma 3n - Text Only (LM)
Gemma 3n
Nanonets OCR
BitNet 1.58
Holo1
Qwen3 DWQ Quants
DeepSeek R1 0528
AceReason Nemotron
Devstral Small 2505
MedGemma
OuteTTS-1.0
SimCLRv1
Gemma 3 DWQ
Parakeet
Josiefied and Abliterated Qwen3
Qwen3
NariLabs Dia-1.5B
GLM4
PLaMo
Kimi-VL Thinking
Gemma 3 QAT
Llama 4
ModernBert
Gemma 3
Qwen QwQ
OLMoE
UI-TARS
olmOCR
Kokoro TTS
FuseAI
Arcee Virtuoso
Mistral Small
DeepSeek-R1-Distill
Qwen2.5-VL
Qwen2.5-1M
Mamba2
Jina Reader-LM
Helium-1
QVQ-72B-Preview
DeepSeek-VL2
Josiefied and Abliterated Qwen2.5
Falcon3 Mamba
EXAONE-3.5
Llama 3.3
Paligemma 2
QwQ-32B-Preview
Idefics 3 + SmolVLM
Molmo
Florence-2
Falcon-Mamba
Qwen2.5-Coder
Ministral
Lumimaid
Mistral
Qwen1.5
Code Gemma
Llama 3
Qwen2.5
Phi-3
Whisper
OpenELM
Yi-1.5
Mistral (Mamba) Codestral
Google Gemma2
Mamba
HF SmolLM
Mistral NeMo
Llama 3.1
EnCodec
Llama 3.2

Qwen-Image-Flash (MLX)

NVIDIA's DMD2 4-step Qwen-Image distill for MLX/MLX-Swift. bf16 = quality reference; int8 = near-lossless, 4x faster, 30 GB peak.