● Covered by 1 source · 1 reportLow impact1 neutral
Jeffy Classify Offers CPU-Runnable Pretrained Text Classifiers with Retraining Capability
Key points
- Pretrained text classifiers run on CPU.
- Supports retraining with custom data.
- Includes 13 classifiers with documented weights and licenses.
- Shared encoder (bge-large-en-v1.5) is downloaded once.
Local Text Classification
Jeffy Classify introduces a set of pretrained text classifiers designed to operate on a CPU. This eliminates the need for a GPU, making text classification accessible on a wider range of hardware configurations. The package includes 13 classifiers, with their weights derived from logistic regression coefficients.
Retraining Capabilities
Users can retrain these classifiers with their own data. The tool supports common data formats like .csv, .tsv, and .jsonl for input. An example retraining process is provided using a bundled 24-row product review CSV, demonstrating how to customize models for specific tasks.
Installation and Usage
Installation is managed via `uv` or `pip`. After installation, users can serve the classifiers locally via an HTTP API or integrate them directly into Python code using the `jeffy.engine` module. The first prediction triggers a download of the shared encoder (bge-large-en-v1.5, approximately 1.2 GB), which is then cached for subsequent use.
Performance and Limitations
Test accuracy results on held-out splits are documented. The tool notes weaknesses in performance for SNLI (65.6%) and tweet_eval_sentiment (66.2%) compared to task-specific models. The Emotion classifier (75.5%) has limited class coverage, and probabilities are uncalibrated.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Primary sources
GitHub cmc_internal/api GitHub github/collect GitHub nicobrenner/jeffy GitHub _private/browser GitHub get-started/accessibility GitHub open-source/sponsorsReporting from
Jeffy Classify, a new tool, provides pretrained text classifiers that can be run and retrained on a CPU without requiring a GPU. This allows for local text classification and model customization on standard hardware.