Open models to fine-tune and serve
Model families
28 families
- Supported
Qwen3
Qwen
- Dense
- Text
- Sizes
- 0.6B
- 1.7B
- 4B
- 8B
- 14B
- 32B
- Suggested sizes
- LoRA from 1 GPU, about 9 GB
- Serves
- Adapter, base model
- Supported
Qwen3 MoE
Qwen
- Mixture of experts
- Text
- Sizes
- 30B, 3B active
- Suggested sizes
- LoRA from 1 GPU, about 78 GB
- Serves
- Adapter, base model
- Supported
Qwen3.5, Qwen3.6 and Qwen3.8
Qwen
- Dense
- Text and images
- Sizes
- 9B
- 27B
- Suggested sizes
- LoRA from 1 GPU, about 71 GB
- Serves
- Adapter, base model
- Supported
Qwen3.5 and Qwen3.6 MoE
Qwen
- Mixture of experts
- Text and images
- Sizes
- 35B, 3B active
- 122B, 10B active, serving only
- Suggested sizes
- LoRA from 1 GPU, about 91 GB
- Serves
- Adapter, base model
- Experimental
Qwen3.8 Flash-Next
Qwen
- Mixture of experts
- Text and images
- Sizes
- 180B
- Suggested sizes
- LoRA from 8 x NVIDIA B200
- Serves
- Adapter, base model
- Supported
gpt-oss
OpenAI
- Mixture of experts
- Text
- Sizes
- 20B
- 120B, serving only
- Suggested sizes
- LoRA from 1 GPU, about 55 GB
- Serves
- Adapter, base model
- Supported
GLM-4.7 Flash
Z.ai
- Mixture of experts
- Text
- Sizes
- 31B
- Suggested sizes
- LoRA from 1 GPU, about 80 GB
- Serves
- Adapter, base model
- Supported
Gemma 4
Google
- Dense and mixture of experts
- Text and images
- Sizes
- E2B
- E4B
- 12B
- 26B, 4B active
- 31B
- Suggested sizes
- LoRA from 1 GPU, about 33 GB
- Serves
- Adapter, base model
- Supported
Laguna 2.1
Poolside
- Mixture of experts
- Text
- Sizes
- 33B
- 118B, serving only
- Suggested sizes
- LoRA from 1 GPU, about 85 GB
- Serves
- Adapter, base model
- Supported
LFM2 MoE
Liquid AI
- Mixture of experts
- Text
- Sizes
- 8B, 1B active
- 24B, 2B active
- Suggested sizes
- LoRA from 1 GPU, about 25 GB
- Serves
- Adapter, base model
- Supported
NVIDIA Nemotron 3.5
NVIDIA
- Mixture of experts
- Text
- Sizes
- 30B, 3B active
- 120B, 12B active, serving only
- Suggested sizes
- LoRA from 1 GPU, about 81 GB
- Serves
- Adapter, base model
- Supported
Llama
Meta
- Dense
- Text
- Sizes
- 1B
- 3B
- 8B
- Suggested sizes
- LoRA from 1 GPU, about 7 GB
- Serves
- Adapter, base model
- Supported
SmolLM2
Hugging Face
- Dense
- Text
- Sizes
- 360M
- 1.7B
- Suggested sizes
- LoRA from 1 GPU, about 8 GB
- Serves
- Adapter, base model
- Supported
Mistral
Mistral AI
- Dense
- Text
- Sizes
- 7B
- Suggested sizes
- LoRA from 1 GPU, about 22 GB
- Serves
- Adapter, base model
- Supported
Phi-3
Microsoft
- Dense
- Text
- Sizes
- 3.8B
- Suggested sizes
- LoRA from 1 GPU, about 13 GB
- Serves
- Adapter, base model
- Supported
DeepSeek R1 distills
DeepSeek
- Dense
- Text
- Sizes
- 1.5B
- 7B
- 8B
- 14B
- 32B
- 70B
- Suggested sizes
- LoRA from 1 GPU, about 8 GB
- Serves
- Adapter, base model
- Supported
DeepSeek V2 Lite
DeepSeek
- Mixture of experts
- Text
- Sizes
- 16B
- Suggested sizes
- LoRA from 1 GPU, about 42 GB
- Serves
- Adapter, base model
- Supported
Qwen2.5-VL
Qwen
- Dense
- Text and images
- Sizes
- 3B
- 7B
- Suggested sizes
- LoRA from 1 GPU, about 13 GB
- Serves
- Adapter, base model
- Supported
Qwen3-VL
Qwen
- Dense
- Text and images
- Sizes
- 4B
- 8B
- Suggested sizes
- LoRA from 1 GPU, about 15 GB
- Serves
- Adapter, base model
- Supported
Gemma 3
Google
- Dense
- Text and images
- Sizes
- 4B
- 12B
- 27B
- Suggested sizes
- LoRA from 1 GPU, about 14 GB
- Serves
- Adapter, base model
- Supported
SmolVLM
Hugging Face
- Dense
- Text and images
- Sizes
- 2B
- Suggested sizes
- LoRA from 1 GPU, about 9 GB
- Serves
- Adapter, base model
- Supported
ModernBERT and Ettin (decision models)
Answer.AI and Johns Hopkins University
- Dense
- Text
- Decision model
- Sizes
- 150M
- 400M
- 396M
- Suggested sizes
- LoRA from 1 GPU, about 7 GB
- Serves
- Adapter, base model
- Supported
Mistral Small 4
Mistral AI
- Mixture of experts
- Text
- Serving only
- Sizes
- 119B, serving only
- LoRA from
- -
- Serves
- Base model
- Supported
Command A+
Cohere
- Mixture of experts
- Text
- Serving only
- Sizes
- 219B, serving only
- LoRA from
- -
- Serves
- Base model
- Supported
Inkling
Thinking Machines
- Mixture of experts
- Text
- Serving only
- Sizes
- 266B, serving only
- LoRA from
- -
- Serves
- Base model
- Supported
GLM-5.3 Flash
Z.ai
- Mixture of experts
- Text
- Serving only
- Sizes
- 321B, serving only
- LoRA from
- -
- Serves
- Base model
- Supported
Step 3.7 Flash
StepFun
- Mixture of experts
- Text
- Serving only
- Sizes
- 201B, serving only
- LoRA from
- -
- Serves
- Base model
- Supported
Qwen3-Next
Qwen
- Mixture of experts
- Text
- Serving only
- Sizes
- 80B, 3B active, serving only
- LoRA from
- -
- Serves
- Base model
Model check
What about other models?
Any other Hugging Face chat model with a chat template is checked when you create a project. A model outside these families passes with a warning: it may not train or serve. See the supported-models docs.
Start with one dataset.
Create an account and import your data, or book a demo and we will show you the product.