test-time-scaling 话题在 GitHub 上没有填写简介，这里按 Star 数汇总最相关的热门仓库。
🚀 MassGen is an open-source multi-agent scaling system that runs in your terminal, autonomously orchestrating frontier models and agents to collaborate, reason, and produce high-quality results. | Join us on Discord: discord.massgen.ai
Inference-time scaling for LLMs-as-a-judge.
[ICCV 2025] Video-T1: Test-Time Scaling for Video Generation
SE-Agent 是一个面向 LLM 代码代理的自我演进框架。它使轨迹级演化能够通过修订、重组和精炼等推理路径交换信息，扩展搜索空间并逃离局部最优。在 SWE-bench Verified 上，它实现了 SOTA 性能
Official codebase for "Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling".
Official code for "Student Guides Teacher: Weak-to-Strong Inference via Spectral Orthogonal Exploration" (SOE, ACL 2026 Oral). A training-free, test-time scaling method that fixes reasoning collapse in LLM mathematical reasoning via orthogonal probing.
Repo of paper "Free Process Rewards without Process Labels"
Probing Scientific General Intelligence of LLMs with Scientist-Aligned Workflows
🔥[ICML'25] Official repository for the paper "Alpha-SQL: Zero-Shot Text-to-SQL using Monte Carlo Tree Search"
"what, how, where, and how well? a survey on test-time scaling in large language models" repository
[AAAI 2026] Official codebase for "GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning".
ACL'2025: SoftCoT: Soft Chain-of-Thought for Efficient Reasoning with LLMs. and preprint: SoftCoT++: Test-Time Scaling with Soft Chain-of-Thought Reasoning
Two Heads are Better Than One: Test-time Scaling of Multi-agent Collaborative Reasoning (EMNLP2026)
Best-of-N coding agents with LLM-as-a-Verifier selection for Oh My Pi: isolated git worktrees, verifier-ranked trajectories, winner-only patch application
A comrephensive collection of learning from rewards in the post-training and test-time scaling of LLMs, with a focus on both reward models and learning strategies across training, inference, and post-inference stages.
[ICML 2026] Decoding Tree Sketching (DTS): a training-free & model agonistic & plug-in framework for LLM parallel reasoning.
Awesome-Parallel-Reasoning: Unlocking the reasoning potential of LLMs. Papers, Code, Resources & Survey.
Inference Engine for SiliconMind-V1 Verilog Coding Models
[ML4H'25] m1: Unleash the Potential of Test-Time Scaling for Medical Reasoning in Large Language Models
A framework for verifiable reasoning with language models.