Zero Shot Classification Python Packages

open-clip-torch

An open source implementation of CLIP.

gutfeel

Judgment calls as one line of Python, built for TypeSafe AI's Jev and running on any small model: likely / classify / rate → YES, NO or UNSURE. Also local NLI, local LLMs, Ollama, vLLM, OpenAI.

logit-classifier

Local zero-shot classifier for text and images. Declares options, returns a calibrated probability for each, generates no text. Accepts the TypeSafe System One request format.

gutfeel-mcp

Judgment calls as one line of Python, built for TypeSafe AI's Jev and running on any small model: likely / classify / rate → YES, NO or UNSURE. Also local NLI, local LLMs, Ollama, vLLM, OpenAI.

discern-vl

Ask a vision model a question, get the answer from its logits in one forward pass. No decoding, nothing trained. The confidence is the logit gap in nats.

brier

Typed, calibrated decisions from any open LLM, read straight from its next-token probabilities.

tachyroute

The blazing-fast, multimodal AI decision engine. Uses adaptive early-exit computing and attention rollout for sub-15ms explainable zero-shot classification

ejai

An 11 MB model that makes typed decisions on CPU (choice, yes/no, score) with calibrated probabilities. No LLM, no tokens.

verdictml

Small, fast, honest decision models. Open alternative to Jev: zero-shot, fit on your labels in seconds, calibrated with a coverage guarantee, Jev wire-compatible.

open-clip-torch-any-py3

An open source implementation of CLIP.

autodistill-gpt-4o

GPT-4o (with Vision) module for use with Autodistill.

decima-ai

Open, CPU-sized Jev-style (System One) decision model: situation + question + options → calibrated probabilities. 122M, ~20 ms on one CPU core, runs in your browser.

autodistill-fastvit

FastViT base model for use with Autodistill.

arbor-process

NeurIPS 2024 Track on Datasets and Benchmarks (Spotlight)

simple-clip

A minimal, but effective implementation of CLIP (Contrastive Language-Image Pretraining) in PyTorch

zero-shot-labeler

A lite zero-shot labeler