bharath-inferock commited on
Commit
ca6c841
·
verified ·
1 Parent(s): 837f589

Results: Dane's own result only (A178)

Browse files

Founder word 11/13 6:0x pm PT: the Results speak only about Dane; the board-leader comparison sentence, the index-leader line and the leader table row are retired. Text of record: DANE-HF-MODEL-CARD-DRAFT-20261113.md (lead workroom).

Files changed (1) hide show
  1. README.md +2 -3
README.md CHANGED
@@ -53,17 +53,16 @@ Dane is a hybrid, frontier class model. It beats the frontier and delivers the m
53
  ## What Dane is
54
 
55
  Dane is a hybrid model made by Inferock.
56
- Dane scores 79.4 on HLE Diamond, 15 percent above the Artificial Analysis board leader.
57
  Dane is API-only: you reach it through the Inferock API as the model id `dane`, and no weights are published on the Hub.
58
 
59
  ## Results
60
 
61
- **HLE Diamond: Dane 79.4, index leader 69.1**
62
 
63
  | Benchmark | Score |
64
  |---|---|
65
  | HLE Diamond | 79.4 |
66
- | Artificial Analysis board leader | 69.1 |
67
 
68
  HLE Diamond is the 1,000-question Diamond set of Humanity's Last Exam (dataset cais/hle-diamond), run with tools and graded as the Artificial Analysis board grades: the official HLE judge prompt, GPT-5.6 Luna (medium) as the judge, pass@1. Score is right over judged: 794 of 1,000; a question with no answer counts as wrong. Run by Inferock on November 4, 2026.
69
 
 
53
  ## What Dane is
54
 
55
  Dane is a hybrid model made by Inferock.
56
+ Dane scores 79.4 on HLE Diamond: 794 of 1,000 questions correct, with tools, run by Inferock on November 4, 2026.
57
  Dane is API-only: you reach it through the Inferock API as the model id `dane`, and no weights are published on the Hub.
58
 
59
  ## Results
60
 
61
+ **HLE Diamond: Dane 79.4 (794 of 1,000)**
62
 
63
  | Benchmark | Score |
64
  |---|---|
65
  | HLE Diamond | 79.4 |
 
66
 
67
  HLE Diamond is the 1,000-question Diamond set of Humanity's Last Exam (dataset cais/hle-diamond), run with tools and graded as the Artificial Analysis board grades: the official HLE judge prompt, GPT-5.6 Luna (medium) as the judge, pass@1. Score is right over judged: 794 of 1,000; a question with no answer counts as wrong. Run by Inferock on November 4, 2026.
68