B1 encoder fine-tuned on ImageNet-1k
According to the model card, this is a b1-sized SegFormer encoder fine-tuned on ImageNet-1k.
Open Source Model Profile · nvidia
mit-b1
mit-b1 is a SegFormer image-classification encoder from nvidia. According to the model card, it is a b1-sized encoder fine-tuned on ImageNet-1k.
- Publisher
- nvidia
- Task
- image-classification
- Model type
- segformer
- License
- other
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
mit-b1 is published by nvidia for image classification. The captured configuration identifies SegformerForImageClassification with model type segformer, and card data records an 'other' license value. According to the model card, it is a b1-sized SegFormer encoder fine-tuned on ImageNet-1k.
Recorded capabilities
Pre-trained encoder for downstream tuning
According to the model card, the SegFormer design pairs a hierarchical Transformer encoder with a lightweight MLP decode head, and this repository holds the pre-trained encoder for fine-tuning use.
Documented 1,000-class inference path
According to the model card, the documented workflow classifies COCO 2017 images into the 1,000 ImageNet classes with SegformerFeatureExtractor and SegformerForImageClassification.
Hugging Face team authorship note
According to the model card disclaimer, the releasing team did not write a card for this model, so the card text was written by the Hugging Face team.
Use cases in the source record
- 1,000-class ImageNet image classification through the publisher-documented Transformers feature-extractor workflow.
- Downstream segmentation or classification fine-tuning starting from the publisher-described pre-trained hierarchical encoder.
Limitations and unknowns
- No Safetensors parameter count was captured for this record.
- No evaluation results were extracted from this record.
- License detail is limited to a publisher link reference, with card data recording an 'other' license value.
- Provider state is historical snapshot data and should be refreshed before being presented as current.
Source and provenance
Source: nvidia/mit-b1
Captured: Unknown. Processed: 2026-09-07T19:34:54.965868+00:00.
SegFormer (b1-sized) encoder pre-trained-only SegFormer encoder fine-tuned on Imagenet-1k. It was introduced in the paper SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers by Xie et al. and first released in this repository . Disclaimer: The team releasing SegFormer did not write a model card for this model so this model card has been written by the Hugging Face team. Model description SegFormer consists of a hierarchical Transformer encoder and a lightweight all-MLP decode head to achieve great results on semantic segmentation benchmarks such as ADE20K and Cityscapes. The hierarchical Transformer is…
F001F002F003F004F005F006F008F009F010F011F012F013F014F015