Unverified50% confidenceBenchmarkExact time
In the creator's leaderboard test, Hydrogen's High level entered the top ten, with a time close to frontier model levels.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
11/15/2026
First Seen
Valid until: 11/29/2026
Sources
Related Entities
Related Claims
UnverifiedArtificial Analysis's model page shows Hydrogen's index scores across four thinking levels: Low 27, Medium 41, High 51, and Ultra 54 (these are index scores on a cross-task suite, not percentages).71% similarUnverifiedIn early testing, partners saw Hydrogen run 2 to 2.8 times faster on complex graph, SQL, and data-intensive workflows, while token usage dropped by 12% to 23%.69% similarUnverifiedWith its chain of thought scaled up, Hydrogen scored 94 in the evaluation, surpassing the previous-generation main flagship Argon.60% similarUnverifiedScale AI的研究团队发现,部分模型在被污染的基准上的表现比在新编写的同等难度题目上高出10-15个百分点59% similarUnverifiedThe study conducted experiments on ten synthetic instances and two dataset-based Gaussian benchmarks, comparing the optimized schedule with the original SR over a finite time horizon.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1172003API
curl https://kongchang.com/api/v1/knowledge/claims/1172003MCP
get_claim(id=1172003)