AI Paper Atlas

Research map / Benchmarks / AIME 2025

AIME 2025 leaderboard

54 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on AIME 2025. Best accuracy: 95 (GPT-5 (high), NeurIPS 2025).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

AIME 2025 · accuracy

28 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
195GPT-5 (high)MathArena: Evaluating LLMs on Uncontaminated Math Competitions
NeurIPS 2025 · Table 1
Agents, code and math
293.33MILES
GPT-OSS-20B
MILES: Modular Instruction Memory with Learnable Selection for Self-Improving LLM Reasoning
NeurIPS 2026 · Table 1
Reasoning and chain of thought
393.3CDG
DeepSeek-R1-8B
Inference Time Optimization with Confidence Dynamics
ICML 2026 · Table 1
Reasoning and chain of thought
488.54Star Elastic 30B (3.6A)
Nemotron Nano v3 30B
Star Elastic: Many-in-One Reasoning LLMs with Efficient Budget Control
ICML 2026 · Table 1
Reasoning and chain of thought
583.33LT early path selection at 2k tokens
Phi-4-Reasoning-Plus
Tracing the Traces: Latent Temporal Signals for Efficient and Accurate Reasoning
ICLR 2026 · Table 2
Reasoning and chain of thought
680BP-Math-32B BAPO
BP-Math-32B (Qwen2.5-Math SFT)
Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping
ICLR 2026 · Table 1
RL from feedback
775.9Seed1.5-Thinking-Enigmata
Seed1.5-Thinking
Enigmata: Scaling Logical Reasoning in Large Language Models with Synthetic Verifiable Puzzles
NeurIPS 2025 · Scaling results table
Reasoning and chain of thought
874.9RCT-4B + RC
Qwen3-4B-Instruct-2507
Reasoning Cache: Continual Improvement Over Long Horizons via Short-Horizon RL
ICML 2026 · Table 1
Reasoning and chain of thought
971.33Adaptive Think
QwQ-32B
Think or Not? Exploring Thinking Efficiency in Large Reasoning Models via an Information-Theoretic Lens
NeurIPS 2025 · Table 1
Reasoning and chain of thought
1070.4Ours (latent QP projection)
Qwen3-14B
Efficient Reasoning via Constrained Optimization in Latent Space
NeurIPS 2026 · Table 1
Reasoning and chain of thought
18 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

AIME 2025 · pass@1

26 papers, each with its best reported number for its own method; higher is better.

#pass@1 ↑MethodPaperTopic
176.7CAMA
Qwen3-32B
CAMA: Enhancing Mathematical Reasoning in Large Language Models with Causal Knowledge
AAAI 2026 · Table 1
Agents, code and math
276.46LED
Qwen3-4B-Thinking
Restoring Exploration after Post-Training: Latent Exploration Decoding for Large Reasoning Models
ICML 2026 · Table 1
Reasoning and chain of thought
376Step-entropy pruning (80% low-entropy)
Qwen3-8B
Making Slow Thinking Faster: Compressing LLM Chain-of-Thought via Step Entropy
ICLR 2026 · Table 1
Reasoning and chain of thought
466.67MindLoom
Qwen3.5-9B
MindLoom: Composing Thought Modes for Frontier-Level Reasoning Data Synthesis
NeurIPS 2026 · Table 1
Agents, code and math
566.6+SAGE-GRPO
Qwen3-8B
Does Your Reasoning Model Implicitly Know When to Stop Thinking?
ICML 2026 · Table 1
Reasoning and chain of thought
658.8SSFT-32B-GFPO
Qwen2.5-32B-Instruct
Training Large Language Models To Reason In Parallel With Global Forking Tokens
ICLR 2026 · Table 1
Reasoning and chain of thought
757.3DeepMath-Omn-1.5B
OpenMath-Nemotron-1.5B
DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning
ICLR 2026 · Table 2
Reasoning and chain of thought
854.3ReTool
DeepSeek-R1-Distill-Qwen-32B
ReTool: Reinforcement Learning for Strategic Tool Use in LLMs
ICLR 2026 · Table 1
Reasoning and chain of thought
953GCPO
DeepSeek-R1-Distill-Qwen-7B
Group Causal Policy Optimization for Post-Training Large Language Models
AAAI 2026 · Table 1
RL from feedback
1048.9G2RPO (lambda=1.5)
DeepSeek-R1-Distill-Qwen-7B
G$^2$RPO: Geometric GRPO; Escaping LLM's Reasoning Rut to Break Accuracy--Entropy Trade-off
ICML 2026 · Table 1
Reasoning and chain of thought
16 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.