Qwen3.8-Max
Benchmarks
#48/134
better than 64% of all scored models
About
No detailed description available for this model.
Benchmark Scores
| Benchmark | Score | Date | Standing |
|---|---|---|---|
|
Workspace-Bench-Lite
general_agent
|
93.46%
|
— |
#2/6
better than 80%
|
|
BFCL-V4
general_agent
|
72.62%
|
— |
#12/33
better than 66%
|
|
MLS-Bench
coding_agent
|
11.11%
|
— |
#5/6
better than 20%
|
|
PaperBench
coding_agent
|
100.00%
|
— |
#1/15
better than 100%
|
|
Vision2Web
vision_language
|
86.88%
|
— |
#4/9
better than 62%
|
|
OSWorld-G
general_agent
|
100.00%
|
— |
#1/6
better than 100%
|
|
BabyVision
vision_language
|
87.36%
|
— |
#7/27
better than 77%
|
|
WorkSpaceBench
general_agent
|
33.93%
|
— |
#9/15
better than 43%
|
|
GDPVal
general_agent
|
89.64%
|
— |
#5/26
better than 84%
|
|
WebArena-Verified
general_agent
|
90.08%
|
— |
#5/11
better than 60%
|
|
OSWorld-Verified
general_agent
|
95.87%
|
— |
#4/31
better than 87%
|
|
OSWorld V2.1
general_agent
|
66.40
|
— |
#6/7
better than 0%
|
|
Agents' Last Exam (CLI)
general_agent
|
24.80
|
— |
#8/8
better than 0%
|
|
Automation-Bench
general_agent
|
74.81%
|
— |
#7/29
better than 79%
|
|
DeepresearchBII
general_agent
|
49.20
|
— |
#7/7
better than 0%
|
|
SWE-bench Pro
coding_agent
|
72.80%
|
— |
#8/81
better than 91%
|
|
OfficeQA Pro
general_agent
|
88.91%
|
— |
#4/17
better than 69%
|
|
WideSearch
general_agent
|
91.39%
|
— |
#6/28
better than 74%
|
|
WeaveBench
vision_language
|
43.14%
|
— |
#3/5
better than 50%
|
|
MLE-bench Lite
coding_agent
|
15.62%
|
— |
#5/6
better than 20%
|
|
TauBench V3 Banking
general_agent
|
100.00%
|
— |
#1/23
better than 100%
|
|
SkillsBench
general_agent
|
83.45%
|
— |
#4/14
better than 77%
|
|
SpreadsheetBench 2
general_agent
|
53.19%
|
— |
#3/6
better than 60%
|
|
JobBench
general_agent
|
66.88%
|
— |
#13/23
better than 45%
|
|
Apex-Agents
general_agent
|
94.55%
|
— |
#3/27
better than 92%
|
|
Terminal-Bench 2.1 (Terminus-2)
coding_agent
|
100.00%
|
— |
#1/54
better than 100%
|
|
NL2Repo-Bench
coding_agent
|
77.47%
|
— |
#5/26
better than 84%
|
|
Harvey Legal Agent
general_agent
|
96.00%
|
— |
#2/7
better than 83%
|
|
DeepSWE
coding_agent
|
90.66%
|
— |
#8/26
better than 72%
|
What is a good Workspace-Bench-Lite score? What is a good BFCL-V4 score? What is a good MLS-Bench score?
No comments yet
Be the first to share your thoughts!