List your company

Long-horizon agentic task on HLE

60PerformanceAggAgent
7.58421.19234.848.408Apr 13, 2026Apr 19, 2026Apr 26, 2026May 3, 2026May 9, 2026May 16, 2026May 23, 2026

Evaluation results

MethodLinks
AggAgentBackbone=MiniMax-M2.5, Aggregation Category=LLM-based Aggregation, Number of Rollouts (K)=82026.0460
Best-of-NBackbone=MiniMax-M2.5, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0458.06
Summary AggregationBackbone=MiniMax-M2.5, Aggregation Category=LLM-based Aggregation, Number of Rollouts (K)=82026.0458.06
Weighted VotingBackbone=MiniMax-M2.5, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0457.42
AggAgentBackbone=Qwen3.5-122B, Aggregation Category=LLM-based Aggregation, Number of Rollouts (K)=82026.0454.19
Solution AggregationBackbone=MiniMax-M2.5, Aggregation Category=LLM-based Aggregation, Number of Rollouts (K)=82026.0454.19
Best-of-NBackbone=Qwen3.5-122B, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0453.55
Majority VotingBackbone=MiniMax-M2.5, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0453.55
Solution AggregationBackbone=Qwen3.5-122B, Aggregation Category=LLM-based Aggregation, Number of Rollouts (K)=82026.0450.97
Summary AggregationBackbone=Qwen3.5-122B, Aggregation Category=LLM-based Aggregation, Number of Rollouts (K)=82026.0450.32
Kimi-K2.5-1TCM=w/o2026.0550.2
Fewest Tool CallsBackbone=MiniMax-M2.5, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0446.45
Pass@1Backbone=MiniMax-M2.5, Aggregation Category=Single Rollout, Number of Rollouts (K)=12026.0445.73
Weighted VotingBackbone=Qwen3.5-122B, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0445.16
Majority VotingBackbone=Qwen3.5-122B, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0443.87
Claude-4.5-OpusCM=w/o2026.0543.4
Fewest Tool CallsBackbone=Qwen3.5-122B, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0441.29
Pass@1Backbone=Qwen3.5-122B, Aggregation Category=Single Rollout, Number of Rollouts (K)=12026.0439.35
SAMBackbone=GLM-4.72026.0538.2
GLM-4.7CM=summary2026.0537.5
AggAgentBackbone=GLM-4.7-Flash, Aggregation Category=LLM-based Aggregation, Number of Rollouts (K)=82026.0437.42
GLM-4.7CM=w/o2026.0537.2
GLM-4.7CM=recent-k2026.0537.2
SAMBackbone=Qwen3.5-35B-A3B2026.0537.2
GLM-4.7CM=discard-tool2026.0536.5
Qwen3.5-35B-A3BCM=discard-tool2026.0536
GPT-5.4CM=w/o2026.0535.2
Qwen3.5-35B-A3BCM=summary2026.0535.2
Summary AggregationBackbone=GLM-4.7-Flash, Aggregation Category=LLM-based Aggregation, Number of Rollouts (K)=82026.0434.84
Qwen3.5-35B-A3BCM=recent-k2026.0534.2
Qwen3.5-35B-A3BCM=w/o2026.0534
Solution AggregationBackbone=GLM-4.7-Flash, Aggregation Category=LLM-based Aggregation, Number of Rollouts (K)=82026.0432.9
Best-of-NBackbone=GLM-4.7-Flash, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0430.97
IterResearcher-30B-A3BCategory=Agent systems2026.0528.8
Weighted VotingBackbone=GLM-4.7-Flash, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0427.74
Fewest Tool CallsBackbone=GLM-4.7-Flash, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0427.1
Majority VotingBackbone=GLM-4.7-Flash, Aggregation Category=Heuristic Aggregation, Number of Rollouts (K)=82026.0426.45
Pass@1Backbone=GLM-4.7-Flash, Aggregation Category=Single Rollout, Number of Rollouts (K)=12026.0425
OpenAI-o3CM=w/o2026.0524.9
WebThinker-32BCategory=Agent systems2026.0515.8
WebSailor-32BCategory=Agent systems2026.059.6