๐ GigaChat 3.5 Reasoning โ a new open-source LLM that thinks before it answers.
It breaks problems into stages, builds a plan, checks intermediate results, and self-corrects. Built on GigaChat 3.5 Ultra, it explores multiple step-by-step reasoning paths for math & coding, using automated verification to reinforce correct answers.
โก๏ธ Proprietary linear attention makes it highly efficient on long contexts, retaining key points without re-matching from scratch. Itโs also token-efficient: uses 37% fewer tokens than DeepSeek V4 Flash Preview on math problems!
๐ Benchmark gains over non-reasoning version:
โข IFBench: 44 โ 77
โข Natural Plan: 64 โ 80
โข LiveCodeBench v6: 56 โ 85
๐ฆ MIT license. Weights on Hugging Face: fp8 | bf16
๐ GigaChat 3.
NeuralZone | AI Apps
@NeuralZone@NeuralZone โ Curated, hand-picked AI tools and services that are actually useful. Buy ads: https://telega.io/c/NeuralZone contact us via @photofixer
369,167 ืื ืืืื
ืคืชื ืืืืืจื 1 ืกืจืืื ืืฆืืจืฃ ืืคืืกื ืื โ ืืืื ืืืคืืืงืฆืืืช ืืืืจื.