classification
Decision trees: learning a prediction by asking a sequence of questions Lesson
A decision tree learns a sequence of feature-based questions that divides examples into groups and assigns a prediction to each final group.
Microsoft launches Decision-1 to score workflow choices without long generated answers News
Microsoft-Decision-1 returns probabilities for bounded workflow decisions through hosted services, while H2O offers a separate downloadable decision model.
Logistic regression: turning evidence into a probability for a yes-or-no decision Lesson
Logistic regression learns a weighted score from labeled examples and turns it into a probability, separating prediction from the threshold used to act.
Liquid AI releases d1 models that choose answers without generating prose News
Liquid AI released two open-weight decision models that return typed answers in one pass, targeting routing, inspection and bounded agent decisions.
OpenAI launches a Decisions API that returns choices instead of prose News
OpenAI’s Decisions API beta returns bounded decisions from text and images at $0.10 per million input tokens, with no output-token charge.
Support vector machines: learning the widest useful boundary Lesson
A support vector machine learns a decision boundary by balancing a wide separation between classes against classification errors, with kernels enabling curved boundaries.
Cloudflare launches Clef models that score decisions without writing an answer News
Cloudflare’s Clef models score developer-defined choices directly, with hosted APIs, open weights, and vendor-reported workflow speed gains.
Selective prediction and abstention: when an AI should decline to decide Lesson
Selective prediction lets a model answer only when its expected reliability clears a threshold, routing uncertain cases to a person, a stronger model, or another check instead of forcing every input into an answer.
Jeff releases a small open model for fast typed decisions on local hardware News
Jeff is an open 0.8B decision-model project that returns calibrated option choices instead of prose, illustrating where a small local model can be useful without claiming frontier reasoning ability.
Tabular foundation models: teaching one model to understand many spreadsheets Lesson
A tabular foundation model is pretrained across many synthetic or real table problems so it can make useful predictions on a new spreadsheet with little or no per-dataset training.
A 25-line Jev parody shows why closed choices are not calibrated decisions News
A local-Qwen demonstration can reproduce Jev's visible closed-choice probability interface, but it does not evaluate the specialized model's claimed decision quality, calibration, or parallel serving.
Discriminative models: AI that chooses instead of writes Lesson
A discriminative model learns to choose among defined answers or estimate their probabilities, often making it faster, cheaper, and easier to check than a model that generates unrestricted text.
Label smoothing: teaching a classifier not to be certain beyond the evidence Lesson
Label smoothing replaces a perfectly one-hot training target with a slightly softened probability distribution, discouraging a classifier from treating every labeled example as proof that all alternatives are impossible.
ROC curves and AUC: the number behind 'the detector works' Lesson
AUC measures how well a detector ranks positives above negatives across every possible threshold, which is why 0.5 means coin flip and why a high AUC can still hide a useless system.
Softmax and Cross-Entropy: How a Model Turns Scores Into a Confident Guess Lesson
Softmax turns a model's raw scores into a probability distribution, and cross-entropy measures how far that distribution is from the true answer; together they are the training target behind almost every classifier and language model.
TypeSafe Python SDK Tool
The public client library for TypeSafe’s typed System One decision API; Jev access itself remains early-access and waitlisted.
Perplexity Decisions API Tool
A documented API for yes/no probabilities, named choices, and ordered scores; its exact launch date and circulated price are unverified in the dossier.
OpenAI Decisions API Tool
Public beta returning predicate probabilities, fixed choices, and ordered scores from text and inline images; GPT-6 Luna at launch.
Microsoft-Decision-1 Tool
Hosted single-pass scoring for bounded workflow choices, available through Microsoft Foundry and OpenRouter.
Liquid d1-3B Tool
Downloadable decision component with documented local loading and named-device measurements for text or image questions; commercial licensing has a revenue threshold.
Kev Tool
Apache-2.0 local Qwen-based models that return typed probability decisions for yes/no, choice, and rating tasks.
Jeff Tool
Open code and downloadable small models for structured option selection, probabilities, and low-latency local decision loops.
Cloudflare Clef and Clef-Flash Tool
Earlier-release roundup: October 1 Apache-2.0 models for scoring structured choices, including routing and classification, rather than generating prose. Published timing setup uses a single H200; consumer-hardware requirements are not established.
Cloudflare Clef Tool
Shipping schema-bound decision models on Workers AI; the larger hosted model costs $0.24 per million input tokens and Flash costs $0.09.