Spaces:
Paused
Paused
Commit History
-
yes
Junzhe Lidba3d2e
-
Merge branch 'main' into tool-changes
VictorLJZe4e9fae -
fixing agent
Junzhe Lifd330d9
-
Set cuda:0 for medrax provider
Adibvafa35e8f3b -
Merge branch 'main' into feature/benchmark
VictorLJZe6e3048 -
fixes
Junzhe Lid38594f
-
revamped benchmarking suite
Junzhe Li89321e2
-
updated benchmarks
Junzhe Lib93ad3f
-
Fix merge conflicts
Adibvafa128b355
-
fixed rexvqa benchmark and added handling for image norm for tools
victorli16278b5
-
partially fixed rexvqa benchmark
victorlic963ad3 -
added concurrency to benchmarking
victorliaa37a55 -
Fix merge conflicts
Adibvafabc86327 -
Setup MedRAX2 for test
Adibvafae116685
-
fixed medgemma
victorli7b3e756
-
cleared merge issues
victorlie97f266 -
adding medgemma
victorli1516987
-
final updates
044eaf7 -
refactored tools
25fe9b8
-
added duckduckgo tool
7608486
-
add cuda
Adibvafa6ee3108 -
updates
358df7a -
updates
148dc3c -
updates
d6cb1b4 -
final fixed version
c90e4b6 -
changes
6db0a72 -
merge
719e93e -
modified agent setup
05f20bf
-
Chagne formatting to use boxed
Adibvafa36f77f2 -
Prep for actual benchmarking
Adibvafa7f4d4c2 -
fixed medrax provider
e25f680 -
updates
587ddab -
Merge branch 'feature/benchmark' into victor
4216843
-
Seed random shuffle
Adibvafa6bbb74f -
Enable shuffling for benchmark
Adibvafaddb8bad
-
Merge branch 'feature/benchmark' into victor
35025a2 -
updates
d01acb6 -
refactored rexvqa benchmark
eaff77c -
Fix prompt load
Adibvafaa7d0aad -
Supoort individual logging
Adibvafa5084d75 -
removed anthropic provider
f350e79
-
added anthropic provider
b0f54ed -
addressed PR comments
9c54b1c -
added openrouter provider
f60c51c -
adding this just in case
e040fb2 -
rexvqa now works
c7b65ec
-
updates
d07a267
-
added chestagentbench
2b26ed4 -
updates
e08f161 -
simplifying evaluations
a8f2960