Model reference · open weights

sapiens2-pretrain

Available as managed deployment Licence fee Embeddings facebook Image embed 3 variants 1k dl/mo

sapiens2-pretrain is an open-weight embedding model from facebook. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.

Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.

What it is

Released byMeta
Published underfacebook
TypeEmbedding models
TaskImage embed
Parameters (lead)395M
Runs withsapiens2
Released2026-04-23
Popularity1k downloads / month
LicenceCommercial licence needed

About

What sapiens2-pretrain is

Sapiens2 is a family of high-resolution vision transformers pretrained on 1 billion human images — designed for human-centric tasks such as pose estimation, body-part segmentation, surface normals, and pointmaps.

This repository contains the 0.4B parameter pretrained backbone. It produces dense per-patch features suitable for fine-tuning downstream task heads.

Read the full model card

Model Details

  • Developed by: Meta
  • Model type: Vision Transformer
  • License: Sapiens2 License
  • Task: pretrain
  • Format: safetensors
  • File: sapiens2_0.4b_pretrain.safetensors

Quick Start

Install the Sapiens2 repo (pip install -e .).

import torch
from huggingface_hub import hf_hub_download
from safetensors.torch import load_file
from sapiens.backbones.standalone.sapiens2 import Sapiens2

# Build the model and load the pretrained checkpoint
model = Sapiens2(arch="sapiens2_0.4b", img_size=(1024, 768), patch_size=16).eval().cuda()  # img_size is (H, W)
ckpt_path = hf_hub_download(repo_id="facebook/sapiens2-pretrain-0.4b", filename="sapiens2_0.4b_pretrain.safetensors")
model.load_state_dict(load_file(ckpt_path))

# Forward pass on a single image (RGB; ImageNet normalization recommended)
x = torch.randn(1, 3, 1024, 768).cuda()
with torch.no_grad():
    features = model(x)[0]  # dense backbone features: (B, num_tokens, embed_dim)

Model Card

FieldValue
ArchitectureSapiens2 ViT (RoPE, GQA, SwiGLU, RMSNorm, QK-norm)
Parameters0.398 B
FLOPs1.260 T
Embedding dim1024
Layers24
Attention heads16
Pretraining resolution1024 × 768 (H × W)
Patch size16
Pretraining data1B human images

Sapiens2 Family

ModelParamsFLOPsEmbed dimLayersHeads
Sapiens2-0.1B0.114 B0.342 T7681212
Sapiens2-0.4B (this)0.398 B1.260 T10242416
Sapiens2-0.8B0.818 B2.592 T12803216
Sapiens2-1B1.462 B4.715 T15364024
Sapiens2-1B-4K1.607 B15364024
Sapiens2-5B5.071 B15.722 T24325632

See the Sapiens2 Collection for all variants and downstream task checkpoints (pose, segmentation, normals, pointmaps).

Intended Use

  • Feature extraction for human-centric downstream tasks
  • Initialization for fine-tuning task heads (pose, segmentation, normals, pointmap)
  • Research on human-centric vision

License

Released under the Sapiens2 License.

Citation

@article{khirodkarsapiens2,
  title={Sapiens2},
  author={Khirodkar, Rawal and Wen, He and Martinez, Julieta and Dong, Yuan and Su, Zhaoen and Saito, Shunsuke},
  journal={arXiv preprint arXiv:2604.21681},
  year={2026}
}

From the published model card. Full card on the HuggingFace links in the sidebar.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys sapiens2-pretrain for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (sapiens2-pretrain below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/embeddings \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"sapiens2-pretrain","input":"text to embed"}'

Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms