SaylorTwift HF Staff commited on
Commit
da8e640
·
verified ·
1 Parent(s): 0f389ab

Add community evaluation results

Browse files

Adds .eval_results/Nex-N2.5-Pro.yaml with results mapped to registered Hub benchmarks (Terminal-Bench 2.1, SWE-bench Pro, DeepSWE v1.1, Toolathlon-Verified), extracted from the Nex-N2.5 model card's Pro-column benchmark table.

Files changed (1) hide show
  1. .eval_results/Nex-N2.5-Pro.yaml +38 -0
.eval_results/Nex-N2.5-Pro.yaml ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ - dataset:
2
+ id: harborframework/terminal-bench-2.1
3
+ task_id: terminalbench_2_1
4
+ value: 82.7
5
+ date: "2026-09-11"
6
+ source:
7
+ url: https://huggingface.co/nex-agi/Nex-N2.5-Pro
8
+ name: "Nex-N2.5 model card"
9
+ notes: "NexAU harness."
10
+
11
+ - dataset:
12
+ id: ScaleAI/SWE-bench_Pro
13
+ task_id: SWE_Bench_Pro
14
+ value: 61.2
15
+ date: "2026-09-11"
16
+ source:
17
+ url: https://huggingface.co/nex-agi/Nex-N2.5-Pro
18
+ name: "Nex-N2.5 model card"
19
+ notes: "NexAU harness."
20
+
21
+ - dataset:
22
+ id: datacurve/deep-swe
23
+ task_id: deep_swe
24
+ value: 55.8
25
+ date: "2026-09-11"
26
+ source:
27
+ url: https://huggingface.co/nex-agi/Nex-N2.5-Pro
28
+ name: "Nex-N2.5 model card"
29
+ notes: "NexAU harness."
30
+
31
+ - dataset:
32
+ id: hkust-nlp/Toolathlon
33
+ task_id: toolathlon_verified
34
+ value: 68.5
35
+ date: "2026-09-11"
36
+ source:
37
+ url: https://huggingface.co/nex-agi/Nex-N2.5-Pro
38
+ name: "Nex-N2.5 model card"