Spaces:
Sleeping
Sleeping
Commit History
-
add missing report file
795a953 -
add CoAct v1.0 trajectories
013500c
-
Add CoAct v1.0 trajectory (#10)
Verified signature- add CoAct traj and fix viz page (c9f7744eee08c5cc8c5e6ca75e32f17edda25d9f) Co-authored-by: Ryan Tran <ryanhoangt@users.noreply.huggingface.co>10afdaa
-
add collapsable
f5de982 -
update instruction
1d7a98e -
remove all the with hint result
d786aec -
cleanup metrics and fix repo
84f76a1 -
cleanup metrics and fix repo
66b7775 -
add llama-3.1 result
fb96108 -
rename file
e3056b0 -
rename OpenDevin to OpenHands
adf5af2
-
bump streamlit ver
85a962a -
fix typo
91ef200 -
update app file
bbf74e2 -
fix visualizer with latest streamlit feature
248fd06 -
add 2nd run
--global455affb
-
add gpt-4o-mini result
Xingyao Wang3d1d4f1
-
Revert "add result from gpt-4o-mini"
Xingyao WangThis reverts commit 3d406f52cf294d6dae0962dfbe2011b998d9224d.12597ea -
add result from gpt-4o-mini
Xingyao Wang3d406f5
-
update the last missing instance
Xingyao Wang1aaf82f -
update result from pr2489
Xingyao Wangfc34a41
-
remove keys
Xingyao Wang77dbd55 -
revoke keys
Xingyao Wanga34dfe3 -
add gpqa result
Xingyao Wang804693c
-
update v1.8 perf
Xingyao Wangec5bc65
-
add result for v1.8 no-hint gpt4o
Xingyao Wangbd3dee6
-
fix model_name in updated metadat
Xingyao Wangdf68ce0 -
add v1.8 result
Xingyao Wangbb84cd4
-
update results using new ver of swebench
Xingyao Wang091b42e
-
set n error/stuck/cost to 0 for CodeAct exp run below v1.5
Xingyao Wangd2b6426 -
by default not showing with hint result
Xingyao Wangba8f82b -
add claude-3.5 result
Xingyao Wang1aa3b7d
-
support loading report with new format
Xingyao Wange2ddd17 -
update gitignore
Xingyao Wang98bdf36 -
update old result w/ swe-bench latest harness;
Xingyao Wangadd v1.7-no-hint result for 4o;68dee1f
-
improved patch apply
Xingyao Wang9071da3 -
improved patch apply
Xingyao Wanga4e8ae8
-
add report field
Xingyao Wang5abf617
-
Add CodeAct 1.6 no hint
Verified signaturef47ed15 -
fix visualizer
Xingyao Wang913979f
-
feat: add gpqa results (#8)
Verified signature- feat: add gpqa results (ea5c515d7c8a0916c9838b9a54eca2f4282712bc) - doc: add reproducibility patch and README for gpqa (a2562f8d900a4598264cf979e7e1ca641d0bcc5e)833a91e
-
fix visualizer to only display eval_report when it exists
Xingyao Wanga4c5e33
-
add result for codeact 1.6
Xingyao Wang03f74db -
only show swe bench on visualizer
Xingyao Wang705a1e5 -
change test_result to bool
Xingyao Wang1ae8615 -
fix fine-grained report; support visualization while running
Xingyao Wang7eb2653 -
add gpt-4-1106 results for codeact swe
Xingyao Wangbb237c5 -
Merge commit 'edc3858a6ea5d0c7317b630024203af60e146b52'
Xingyao Wangf55ef7f -
update all swebench lite
Xingyao Wang78d8859 -
Update outputs/miniwob/README.md
Verified signatureedc3858