Model reference · open weights
vit_pe_core_gigantic_patch14_448.fb is an open-weight embedding model from timm. 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 by | timm |
|---|---|
| Type | Embedding models |
| Task | Image embed |
| Parameters (lead) | 1.9B |
| Runs with | timm |
| Released | 2025-06-20 |
| Popularity | 1k downloads / month |
| Licence | Open weights |
About
This is a timm remapped, image encoder only variant of the original weights.
Perception Encoder (PE) is a state-of-the-art encoder for image and video understanding trained via simple vision-language learning. It was introduced in "Perception Encoder: The best visual embeddings are not at the output of the network".
Model Developer: Meta
Model Overview: Perception Encoder (PE) is a family of large-scale vision encoder models with state-of-the-art performance on a large variety of vision tasks. By using a robust contrastive pretraining recipe and finetuning on synthetically aligned videos, PE not only outperforms all existing models on classification and retrieval, but it also internally produces strong, general features that scale for downstream tasks. PE unlocks the ability for large-scale contrastive pretraining to transfer to downstream tasks with alignment tuning to capitalize on those general features.
PE core is our base model trained with our robust image pretraining schedule and finetuned on the data generated by our synthetic video data engine.
PE core curently comes in 3 sizes. PE core G is the main checkpoint, with L and B models distilled from it.
| Scale | Tower | Params | Width | Depth | MLP | Heads | CLIP Dim | Resolution / Context Len |
|---|---|---|---|---|---|---|---|---|
| B/16 | Vision | 0.09B | 768 | 12 | 3072 | 12 | 1024 | 224px |
| Text | 0.31B | 1024 | 24 | 4096 | 16 | 1024 | 32 tokens | |
| L/14 | Vision | 0.32B | 1024 | 24 | 4096 | 16 | 1024 | 336px |
| Text | 0.31B | 1024 | 24 | 4096 | 16 | 1024 | 32 tokens | |
| G/14 | Vision | 1.88B | 1536 | 50 | 8960 | 16 | 1280 | 448px |
| Text | 0.47B | 1280 | 24 | 5120 | 20 | 1280 | 72 tokens |
All PE core models use an attention pooling block with 8 heads on top of the vision tower. The L and B models additionally have a class token for global aggregation. See the paper for more details.
PE core obtains extremely strong results across the board on zero-shot image classification and retrieval as well as zero-shot video classification and retrieval. We present a sample of its performance across those domains below.
| Model | Checkpoint | IN-1k | IN-v2 | IN-A | ObjectNet | COCO-T2I | Kinetics-400 | VTT-T2I |
|---|---|---|---|---|---|---|---|---|
| B/16 224px | PE-Core-B16-224 | 78.4 | 71.7 | 62.4 | 71.9 | 50.9 | 65.6 | 47.6 |
| L/14 336px | PE-Core-L14-336 | 83.5 | 77.9 | 89.0 | 84.7 | 57.1 | 73.4 | 50.3 |
| G/14 448px | PE-Core-G14-448 | 85.4 | 80.2 | 92.6 | 88.2 | 58.1 | 76.9 | 51.2 |
PE core performs particularly well on the hard benchmarks such as ObjectNet and ImageNet-A.
If you find our code useful for your research, please consider citing:
@article{bolya2025PerceptionEncoder,
title={Perception Encoder: The best visual embeddings are not at the output of the network},
author={Daniel Bolya and Po-Yao Huang and Peize Sun and Jang Hyun Cho and Andrea Madotto and Chen Wei and Tengyu Ma and Jiale Zhi and Jathushan Rajasegaran and Hanoona Rasheed and Junke Wang and Marco Monteiro and Hu Xu and Shiyu Dong and Nikhila Ravi and Daniel Li and Piotr Doll{\'a}r and Christoph Feichtenhofer},
journal={arXiv},
year={2025}
}
@article{cho2025PerceptionLM,
title={PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding},
author={Jang Hyun Cho and Andrea Madotto and Effrosyni Mavroudi and Triantafyllos Afouras and Tushar Nagarajan and Muhammad Maaz and Yale Song and Tengyu Ma and Shuming Hu and Hanoona Rasheed and Peize Sun and Po-Yao Huang and Daniel Bolya and Suyog Jain and Miguel Martin and Huiyu Wang and Nikhila Ravi and Shashank Jain and Temmy Stark and Shane Moon and Babak Damavandi and Vivian Lee and Andrew Westbury and Salman Khan and Philipp Kr\"{a}henb\"{u}hl and Piotr Doll{\'a}r and Lorenzo Torresani and Kristen Grauman and Christoph Feichtenhofer},
journal={arXiv},
year={2025}
}
From the published model card. Full card on the HuggingFace links in the sidebar.
How it works
Using it via the API
Once AxForge deploys vit-pe-core-gigantic-patch14-448-fb for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (vit-pe-core-gigantic-patch14-448-fb 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":"vit-pe-core-gigantic-patch14-448-fb","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.