# Weibo researchers claim 3B-parameter model matches larger AI systems on demanding math benchmarks

_Wednesday, June 17, 2026 at 4:01 AM EDT · AI, Tech & Business · Latest_

![Weibo researchers claim 3B-parameter model matches larger AI systems on demanding math benchmarks — Primary](https://images.ctfassets.net/jdtwqhzvc2n1/3nN597gjeIB7cuVDi13fi4/27ec6c6a739b9a205e7f3a8ecf4d9ddb/Nuneybits_Vector_art_of_a_laptop-sized_AI_casting_a_giant_geome_e2031be7-1e11-4da1-b7bc-671e6a3ca75e.webp?w=800&q=75)

Researchers at Sina Weibo posted a technical report describing VibeThinker-3B, a language model with 3 billion parameters.

The model scored 94.3 on the AIME 2026 math competition, comparable to much larger systems such as DeepSeek V3.2. A test-time scaling technique raised the score to 97.1.

The paper drew rapid attention on Hugging Face and GitHub but also skepticism on social media regarding benchmark reliability.

## Sources

- [VentureBeat](https://venturebeat.com/technology/why-weibos-tiny-vibethinker-3b-has-the-ai-world-arguing-over-benchmarks-again)

---
Canonical: https://techandbusiness.org/newswire/abbQcWyTjHHxE5QRekncpM
Retrieved: 2026-06-17T10:54:33.309Z
Publisher: Tech & Business (techandbusiness.org)