Nicheloom visitors: 14,106 all-time · 8,505 in the last 7 days

The opportunity tracker for new startups.

Today: 1,383 launchs · Niche of the day: document and PDF processing tools · See full report →

ej

An 11MB local model for typed decisions in one pass

Get picks like this daily. The day's top launches, AI/tech news, and a weekly opportunity spotlight — straight to your inbox.

This is 1 of 225 launches in machine learning developer tools — see how it stacks up on momentum and crowding →

763 other launches read as similar to this one →

Details

External ID
1275189
Source
PH
Company
—
Product
ej
Website domain
producthunt.com
Launched
Oct. 10, 2026
Cohort
—
Upvotes
64
Upvotes percentile
0.97
Tags
Open Source, Artificial Intelligence, GitHub

Description

I built ej for a narrower problem than general classification: given some state and a set of decision questions whose options can be defined at runtime, return a probability distribution for each question in one pass. It handles choice, yes/no, and ordered scores, runs on CPU, needs no token generation, and ships as one 11.4 MB model file. The repo includes training, evaluation, adaptation, packed weights, and the benchmark harness.

Enrichment

Niche
machine learning developer tools
Vertical
Horizontal
Function
Model & infra
Audience
Developer
AI stance
AI-native
Project type
Hobby / open-source project
Normalized one-liner
lightweight local model for fast classification
Manually corrected
False

Could you build this?

No Creating a custom 11MB lightweight machine learning model architecture that performs zero-shot or runtime-defined multi-attribute decisions in a single pass requires novel ML research and custom model training.

What it would actually take: Requires designing a specialized compact neural network architecture (e.g., custom transformer or contrastive encoder), curating a massive generalized dataset, running distributed PyTorch training, and quantizing/compiling weights for fast CPU inference. Requires deep machine learning research and training expertise rather than standard prompt engineering.

Competitors

Other products that read as similar to this one — 763 launches clear the similarity bar, closest 8 shown.

Attention rank: #49 of 764 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 344 days after the earliest competitor.

Other launches for this product