Nicheloom

Market intelligence for builders — see what's gaining traction before it's crowded.

JEV-CPU

Run SemIf (Jev-style semantic-if decisions) on a CPU — no GPU. Reads typed option probabilities straight from an open model in one forward pass, plus a web UI.

Details

External ID
1376904097
Source
GITHUB
Company
—
Product
JEV-CPU
Website domain
github.io
Launched
Sept. 19, 2026
Cohort
—
Upvotes
18
Upvotes percentile
0.5655905713553676
Tags
cpu, decision-making, jev, llm, nli, qwen, semantic-if, semif, transformers, web-ui, zero-shot-classification
Fetched at
Sept. 23, 2026, 5:02 p.m.
Updated at
Sept. 23, 2026, 5:02 p.m.

Enrichment

Theme
local AI inference and runtimes
Vertical
Horizontal
Function
Model & infra
Audience
Developer
AI stance
AI-native
Project type
Hobby / open-source project
Normalized one-liner
decision inference engine for cpu-based language models
Manually corrected
False

Could you build this?

Partial The web interface and application wrapper are easily built, but extracting typed option probabilities in a single CPU forward pass requires deep technical knowledge of transformer logit outputs and vocabulary projection.

What it would actually take: A functional implementation requires binding a lightweight CPU runtime like GGML/llama.cpp to a Python or Rust server. The complex piece is modifying the transformer decoding pipeline to extract and normalize logits across discrete choice tokens in a single pass without performing autoregressive rollouts. This requires an ML engineer experienced in transformer logit mechanics, CPU SIMD vectorization, and constrained sampling.

Competitors

Other products that read as similar to this one — 191 launches clear the similarity bar, closest 8 shown.

Attention rank: #128 of 192 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 171 days after the earliest competitor.

Other launches for this product

Same idea, different domain

Nobody's really built a model & infra tool for Fintech yet.