Skip to content
← Home

Claim Splitter

Paste a claim. A tiny model takes it apart: premises, hidden assumptions, weak spots. It runs on your device.

Model not loaded. Downloaded once, then cached in your browser.

Try

English only · 1200 chars max

How it works

Model
Gemma 3 270M, fine-tuned to decompose claims (full fine-tune, then LoRA).
Data
About 1,400 synthetic examples, including refusals for non-English and off-topic input.
Runtime
4-bit GGUF (253 MB) run by llama.cpp compiled to WebAssembly, on the GPU through WebGPU where the browser has it. Greedy decoding.
Score
Section-header F1 against reference answers went from 0.42 to 0.91 on 68 held-out prompts.

It is small and it is wrong sometimes. Read the output as a sketch of a claim's structure, not a verdict.