WeiboAI / VibeThinker

What it does

• Ultra-Efficient Frontier-Level Reasoning. With only 3B parameters, VibeThinker-3B approaches the performance range of much larger frontier reasoning systems. It matches or closely trails models that are orders of magnitude larger on challen
• Outstanding Capabilities Across Benchmarks. VibeThinker-3B delivers strong and balanced performance across mathematics, coding, and out-of-distribution evaluation. It achieves 94.3 on AIME26, 89.3 on HMMT25, 80.2 Pass@1 on LiveCodeBench v6,
• Upgraded SSP Training Paradigm. VibeThinker-3B systematically upgrades the Spectrum-to-Signal Principle introduced in VibeThinker-1.5B. The post-training pipeline strengthens data synthesis, quality filtering, and curriculum learning in SFT
• Inference-Time Scaling with CLR. VibeThinker-3B introduces Claim-Level Reliability Assessment (CLR), a test-time scaling strategy for answer-verifiable reasoning. CLR further boosts performance on math benchmarks, raising AIME26 from 94.3 t
• Ultra-Efficient. VibeThinker-1.5B redefines the efficiency frontier for reasoning models, achieving state-of-the-art performance in mathematical and coding tasks with only 1.5B parameters, 100x to 600x smaller than giants like Kimi K2 (1000