AI Paper Atlas

Research map / Benchmarks / GPQA-Diamond

GPQA-Diamond leaderboard

71 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on GPQA-Diamond. Best accuracy: 83 (Team (Reveal Expert), ICML 2026).

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.

GPQA-Diamond · accuracy

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

#accuracy ↑MethodPaperTopic
183Team (Reveal Expert)
heterogeneous 4-model team
Multi-Agent Teams Hold Experts Back
ICML 2026 · Table 2
LLM agents
279.1Optimal strategy guidance
Reasoning model
The CoT Encyclopedia: Analyzing, Predicting, and Controlling how a Reasoning Model will Think
ICLR 2026 · Figure 4/text
Reasoning and chain of thought
378.1Seed1.5-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
475.8Pangu Pro MoE w/ TrimR
Pangu Pro MoE
TrimR: Verifier-based Training-Free Thinking Trimming for Efficient Test-Time Scaling
ICLR 2026 · Table 2
Reasoning and chain of thought
574.24MILES
GPT-OSS-20B
MILES: Modular Instruction Memory with Learnable Selection for Self-Improving LLM Reasoning
NeurIPS 2026 · Table 1
Reasoning and chain of thought
674.2LIMO-V2-P
Qwen2.5-32B-Instruct
LIMOPro: Reasoning Refinement for Efficient and Effective Test-time Scaling
NeurIPS 2025 · Table 2
Reasoning and chain of thought
773.7AutoTool
Qwen3-8B
AutoTool: Dynamic Tool Selection and Integration for Agentic Reasoning
ICML 2026 · Table 2
Reasoning and chain of thought
873.1Full reasoning trace (decile 100)
gpt-oss-120b
Probing the Trajectories of Reasoning Traces in Large Language Models
NeurIPS 2026 · Table A2
Reasoning and chain of thought
972.9LASE
GPT-4.1-mini
Efficient Multi-Agent Reasoning via Confidence-Guided Adaptive Debate
ICML 2026 · Table 1
LLM agents
1072.22LT 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
43 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

GPQA-Diamond · pass@1

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

#pass@1 ↑MethodPaperTopic
179.2CoNL
Qwen3-8B
Conversation for Non-verifiable Learning: Self-Evolving Large Language Models through Meta-Evaluation
ICML 2026 · Table 3
Agents, code and math
269.6Nemotron-Cascade-14B-ThinkingNemotron-Cascade: Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Models
NeurIPS 2026 · Table 1
Reasoning and chain of thought
368.2DyCon (Ours)
Qwen3-4B-Thinking-2507
DyCon: Dynamic Reasoning Control via Evolving Difficulty Modeling
ICML 2026 · Table 2
Reasoning and chain of thought
467.67Soft (Gumbel)
QwQ-32B
LLMs are Single-threaded Reasoners: Demystifying the Working Mechanism of Soft Thinking
ICLR 2026 · Table 2
Reasoning and chain of thought
567.3NaturalReasoning, 100K
Llama-3.3-70B-Instruct
NaturalReasoning: Reasoning in the Wild with 2.8M Challenging Questions
NeurIPS 2025 · Table: Pass@1 of Llama-3.3-70B
Agents, code and math
667.2CRD-30B-A3B
Qwen3-30B-A3B
Collaborative Reasoning Distillation via Cross-Feedback and Coherent Curation
NeurIPS 2026 · Table 1
Reasoning and chain of thought
766.79ProRL + GooseReason-0.7M (+270 steps)
Qwen3-4B-Instruct
Golden Goose: A Simple Trick to Synthesize Unlimited RLVR Tasks from Unverifiable Internet Text
ICML 2026 · Table 3
Reasoning and chain of thought
864.1Adaptive control
QwQ-32B
Controlling Thinking Speed in Reasoning Models
NeurIPS 2025 · Adaptive control results table
Reasoning and chain of thought
960.7Multiverse-32B
Qwen2.5-32B-Instruct
Multiverse: Your Language Models Secretly Decide How to Parallelize and Merge Generation
NeurIPS 2025 · Table: Performance comparison
Language, safety and interpretability
1060.39SSFT-32B
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
9 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.