CompanyBench

Billing Write-Off Pack + ACATS Transfer Compliance — Jira + Confluence + SQL + SlackCompliance & Regulatory

13 verifiers

EnvironmentsConfluenceJiraSlackZeta3 SQL

TaxonomyCompliance & RegulatoryCompliance & Regulatory

How each model performed on this task. Every chart states its N, and the reward carries a 95% confidence interval once there is more than one trial to measure.

Core metrics

Check Pass %

Mean share of the task's checks passing, averaged over trials. N = 10 trials per model.

  1. 89%

    GPT-5.5

  2. 86%

    Gemini 3.1 Pro

  3. 61%

    Claude Opus 4.8

  4. 0%

    Gemini 3.5 Flash

Extended metrics

Median Time

Wall-clock minutes, median across trials. N = 10 trials per model.

  1. 11m

    Gemini 3.1 Pro

  2. 5m

    Claude Opus 4.8

  3. 4m

    GPT-5.5

  4. 4m

    Gemini 3.5 Flash

Tool Calls

Tool invocations, median across trials — counted from the trajectories. N = 10 trials per model.

  1. 80

    Gemini 3.1 Pro

  2. 71

    GPT-5.5

  3. 55

    Claude Opus 4.8

  4. 21

    Gemini 3.5 Flash

Verifier × model pass rate

How often each verifier passed, per model. Green is passing, red is failing — a red band across a row is a verifier nothing clears.

VerifierGPT-5.5n=10Claude Opus 4.8n=10Gemini 3.1 Pron=10Gemini 3.5 Flashn=10
checks (13)
deliverable_slack_channel_exists100%100%100%0%
slack_value_billing_trigger_key90%30%80%0%
slack_value_billing_trigger_status90%30%80%0%
slack_value_forwarding_days50%0%0%0%
slack_value_grace_breach_count80%50%100%0%
slack_value_grace_breach_share80%50%100%0%
slack_value_grace_days100%50%100%0%
slack_value_pin_pct100%100%100%0%
slack_value_total_writeoff100%100%100%0%
slack_value_writeoff_state100%100%100%0%
floor_source_and_sql0%0%0%0%
floor_write_back100%100%100%0%
process_floor100%100%100%60%

Failure modes on this task

37 tagged verifier failures, by the stage they were assigned to. Higher is worse.

Failure stageGPT-5.5Claude Opus 4.8Gemini 3.1 ProGemini 3.5 Flash
Answer1 mode7101010