Hugging Face's logo
keras 's Collections
Qwen AgentWorld
Mxbai Embedding
Magistral
BLIP2
MetaCLIP2
Qwen3.5
Qwen3.6
Swin Transformer
Sentence Transformers
Multilingual E5
Qwen3.6 MoE
Qwen3.5 MoE
Harrier Embedding
BGE Embedding
DeiT
Qwen 3
HGNetV2
DINOV2
Qwen 3 MoE
D-FINE
ESM-2
PARSeq
Depth Anything
MedSigLIP
Qwen 3 Embedding
GPT OSS
GPT OSS Safeguard
SmolLM3
DINOV3
Qwen 2.5 Coder
SAM 3
RWKV7
Qwen3-Coder
Qwen2.5-Math
VideoPrism
Moonshine
Xception
Mixtral
Qwen 2.5
Qwen 1.5 MoE
RetinaNet
CLIP
SegFormer
ViT
Phi-3
EfficientNetV2
BASNet
CSPDarkNet
MobileNetV3
SigLIP
PaliGemma 2
Llama 3
StableDiffusion 3.5
EfficientNet
ResNetVD
XLM-RoBERTa
T5
StableDiffusion3
RoBERTa
OPT
Mistral
Vicuna
Llama 2
GPT-2
Falcon
FNet
ELECTRA
DistilBERT
DeBERTaV3
Whisper
VGG
SAM
ResNetV2
ResNetV1
Gemma 2
PaliGemma
DenseNet
MiT
DeepLabV3Plus
Gemma
CodeGemma
BART
BERT
Bloom
ALBERT

Qwen3.5 MoE

Highly efficient Mixture-of-Experts (MoE) large language model for advanced reasoning and text generation.