Skip to content

EthenEthenEthen

LoginPlatformTry Ethen

Open Source Model Profile · typeform

mobilebert-uncased-mnli

mobilebert-uncased-mnli is a compact MobileBERT zero-shot-classification model from typeform. According to the model card, it is an MNLI fine-tuned version of uncased MobileBERT.

Publisher
typeform
Task
zero-shot-classification
Model type
mobilebert
License
Unknown
Library
transformers
Publication status
Accepted · not indexed

Model overview

mobilebert-uncased-mnli is published by typeform as a zero-shot-classification model. The captured configuration identifies MobileBertForSequenceClassification with model type mobilebert, and Safetensors metadata reports about 24.6M parameters. According to the model card, it is the MNLI fine-tuned version of the uncased MobileBERT model.

Recorded capabilities

Compact MobileBERT base

According to the model card, the starting point is the uncased MobileBERT model presented as a compact task-agnostic BERT for resource-limited devices.

MNLI fine-tuning

According to the model card, this version was fine-tuned for Multi-Genre Natural Language Inference, with the multi_nli dataset card cited for detail.

Zero-shot classification use

According to the model card, direct use is zero-shot classification in English.

Use cases in the source record

  • English zero-shot classification workflows using the documented direct-use path.
  • NLI-grounded classification research referencing the cited multi_nli dataset card.

Limitations and unknowns

  • No license value was extracted; the card lists license as needing more information.
  • No context-window value was extracted from this record.
  • No evaluation results were extracted from this record.
  • The card records preprocessing, training, hardware, and technical specifications as needing more information.
  • Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: typeform/mobilebert-uncased-mnli

Captured: Unknown. Processed: 2026-09-07T19:35:00.310452+00:00.

Model Card for MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices Model Details Model Description This model is the Multi-Genre Natural Language Inference (MNLI) fine-turned version of the uncased MobileBERT model . Developed by: Typeform Shared by [Optional]: Typeform Model type: Zero-Shot-Classification Language(s) (NLP): English License: More information needed Parent Model: uncased MobileBERT model . Resources for more information: More information needed Uses Direct Use This model can be used for the task of zero-shot classification Downstream Use [Optional] More information needed. Out-of-Scope Use The model…

F001F002F003F004F005F006F007F009F010F011F012F013F014F015F017