DeepSeek V4 Pro (max)

Rank #3 of 17 Task score 63.35 Resolved 64% of tasks Harness OpenCode Effort max Total cost $11.78 Mean time 18 min Mean tool calls 78

polling

71% of tasks touching this surface resolved

writeback

71% of tasks touching this surface resolved

webhooks

36% of tasks touching this surface resolved

Published attempts

10 of 50 scored tasks released

One official run per task. Click a row for the full trajectory: every tool call, vendor request, graded check and the diff.

Task	Ticket	Score	Checks	Tool calls	Time	Cost	Outcome
task-0001	Sync StaffLine candidates, jobs, applications, and notes into our canonical store	92	29/30	89 (8 failed)	22 min	$0.25	resolved
task-0011	TalentForge connector: consume webhooks while pushing corrections upstream	0	55/59	103 (8 failed)	27 min	$0.39	mandatory fail
task-0015	Bulk-import a migration batch into StaffLine	0	19/25	89 (7 failed)	22 min	$0.38	mandatory fail
task-0016	HireWire connector: push stage-change events + keep an incremental poll fresh	92	45/48	88 (2 failed)	18 min	$0.26	resolved
task-0020	Sync GlobalHire candidates into our canonical store (offset polling)	100	34/34	67 (12 failed)	13 min	$0.21	resolved
task-0025	Reset the CrewCall roster watermark after the tenant rebuild	0	500/503	56 (4 failed)	10 min	$0.13	partial
task-0037	Fix the Placemint redeployment sync	100	306/306	81 (4 failed)	15 min	$0.15	resolved
task-0042	Move the rota mirror onto instant storage	100	117/117	66 (3 failed)	12 min	$0.18	resolved
task-0047	Reconcile and safely apply the mobility stage repair queue	100	100/100	88 (7 failed)	19 min	$0.26	resolved
task-0049	Placemint connector: webhooks, polling, and writeback for a high-volume tenant	100	64/64	103 (5 failed)	33 min	$0.62	resolved