Skip to content
  1. Home
  2. Benchmarks
  3. Integration Bench
  4. Tencent HY 4 Preview

Tencent HY 4 Preview

Integration capability of coding agents on Tencent HY 4 Preview.

Rank #5 of 20 Pass 1 62.62 ±5.36 Resolved 64% of tasks Harness OpenCode Effort max Total cost $28.98 Mean time 31 min Mean tool calls 95 Graded 50/50
66
polling
67% of tasks touching this surface resolved
56
writeback
57% of tasks touching this surface resolved
26
webhooks
27% of tasks touching this surface resolved

Published attempts

50 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.

TaskTicketScoreChecksTool callsTimeCostOutcome
task-0001Sync StaffLine candidates, jobs, applications, and notes into our canonical store9229/3086 (3 failed)26 min$0.52resolved
task-0002Legal-hold export: TalentForge custodians and everything on file against them9762/646020 min$0.46resolved
task-0003Build the weekly Placemint contact sweep100343/34375 (5 failed)22 min$0.40resolved
task-0004Merge StaffLine application stage with Placemint placement stage7750/65143 (5 failed)41 min$0.84resolved
task-0005Keep TalentForge candidates fresh: webhooks + reconciliation polling9424/27149 (5 failed)61 min$0.94resolved timed out
task-0006The case-note mirror is short every morning100159/15982 (1 failed)32 min$0.40resolved
task-0007Harden the Vettly sync connector for production scheduling10039/39104 (4 failed)47 min$0.98resolved
task-0008Build a StaffLine full-sync + writeback connector032/339830 min$0.55mandatory fail
task-0009Migrate off the legacy Bullpen app-token integration10040/4011543 min$0.62resolved
task-0010Build the GlobalHire audit-period extract Compliance keeps asking us for10081/8184 (4 failed)19 min$0.38resolved
task-0011TalentForge connector: consume webhooks while pushing corrections upstream056/59154 (4 failed)57 min$1.24mandatory fail
task-0012Build the full TalentLoop connector: polling, webhooks, and writeback94120/123113 (2 failed)45 min$0.86resolved
task-0013The warehouse and RecruitOS have been out of step for months042/96246 min$0.08partial
task-0014HireWire: clear the stale-screening correction backlog, exactly once09/14116 (4 failed)51 min$0.89mandatory fail
task-0015Bulk-import a migration batch into StaffLine8623/2662 (2 failed)19 min$0.33resolved
task-0016HireWire connector: push stage-change events + keep an incremental poll fresh9245/4897 (3 failed)30 min$0.61resolved
task-0017Move the Brightmoor candidate mirror onto TalentForge's event subscription100160/16065 (2 failed)20 min$0.31resolved
task-0018Rebuild the nightly requisition rollup on the RecruitOS API100142/14265 (6 failed)19 min$0.34resolved
task-0019Rebuild the Vettly change feed for the vaultd cutover100183/18465 (6 failed)20 min$0.43resolved
task-0020Sync GlobalHire candidates into our canonical store (offset polling)10034/34124 (5 failed)59 min$0.77resolved
task-0021Sync GlobalHire candidates, placements, and agencies into our canonical store10021/2181 (1 failed)21 min$0.42resolved
task-0022Harden the GlobalHire backfill for a production tenant026/27157 (7 failed)64 min$0.96mandatory fail
task-0023TalentLoop connector: keep a candidate/application mirror deletion-accurate018/26130 (1 failed)47 min$0.91mandatory fail
task-0024TalentLoop connector: unified freshness across webhook-covered and poll-only entities025/40123 (5 failed)41 min$0.84mandatory fail
task-0025Reset the CrewCall roster watermark after the tenant rebuild0500/503117 (1 failed)30 min$0.54partial
task-0026Push the Q1 close back into Placemint0269/26983 (1 failed)23 min$0.39partial
task-0027Reliability: Interviewly connector loses and regresses updates021/2888 (3 failed)28 min$0.56mandatory fail
task-0028Harden the CrewCall snapshot connector for production10044/44111 (1 failed)36 min$0.61resolved
task-0029Get the Riverside signup file onto the CrewCall roster100102/10283 (2 failed)23 min$0.43resolved
task-0030Census the desk board against the Placemint placement book100325/32583 (2 failed)16 min$0.34resolved
task-0031Rebuild the morning bookable-crew board100101/10159 (5 failed)14 min$0.25resolved
task-0032Build the weekly capability index100116/11863 (5 failed)15 min$0.30resolved
task-0033Keep the RecruitOS pipeline and the Placemint ledger in step100153/153104 (4 failed)29 min$0.64resolved
task-0034Weekly crew top-up is putting people on the gate twice100165/16582 (2 failed)24 min$0.46resolved
task-0035Sync Vettly subjects, checks, and reports into our canonical store10028/28106 (3 failed)32 min$0.55resolved
task-0036Harden the Placemint book extract for the larger tenant0211/42087 (1 failed)32 min$0.71partial
task-0037Fix the Placemint redeployment sync100306/30679 (4 failed)25 min$0.56resolved
task-0038TalentLoop connector: make webhook-driven freshness durable against a misbehaving receiver022/39135 (5 failed)51 min$0.98mandatory fail
task-0039Build the nightly payroll bridge for the punch export10094/94101 (2 failed)22 min$0.50resolved
task-0040Push interview reschedules upstream into Interviewly016/19160 (2 failed)45 min$1.01mandatory fail
task-0041Paygrade connector: employee/assignment sync and writeback030/3883 (4 failed)28 min$0.52mandatory fail
task-0042Move the rota mirror onto instant storage100117/1175718 min$0.42resolved
task-0043The rota warehouse is counting the same night more than once10089/8998 (2 failed)21 min$0.45resolved
task-0044Nightly roster census disagrees with CrewCall about who works here10067/677520 min$0.36resolved
task-0045Replatform the ATS integration: StaffLine to Bullpen v2020/25126 (3 failed)35 min$0.81mandatory fail
task-0046Get the scorecard export into TalentLoop088/896724 min$0.47partial
task-0047Reconcile and safely apply the mobility stage repair queue098/1008924 min$0.62partial
task-0048Migrate the Brackett closure archive onto Paygrade100211/21184 (7 failed)20 min$0.43resolved
task-0049Placemint connector: webhooks, polling, and writeback for a high-volume tenant066/6886 (3 failed)38 min$0.57mandatory fail
task-0050Fix the Placemint fee-corrections connector100254/25474 (2 failed)23 min$0.39resolved