Verdict
Quality matched on 100% of cases and the candidate is cheaper. Latency is informational during calibration.
Cases evaluated
Evaluation complete
Quality matches
100%
Est. monthly savings
Projected from matching events observed over 30 days
Status
The experiment finished and its verdict is available.
Controlled replay compares both models from the same worker. Production latency is shown separately as context and is not subtracted from candidate replay latency.
claude-sonnet-4-6 and claude-haiku-4-5 received the same frozen prompts in alternating order.
Replayed source p50
1,419 ms
claude-sonnet-4-6
Replayed candidate p50
626 ms
claude-haiku-4-5
Paired median delta
-762 ms (-55%)
95% interval: -874 ms to -690 ms
Controlled p90
1,624 ms / 735 ms
Source / candidate
Production source context
1,364 ms p50, 3,576 ms p90
30 captured measurements from the user's environment
Controlled coverage
30 successful pairs
Only complete source/candidate pairs enter the comparison (provider-completion-v1)
The evaluation compares structured content canonically and reports presentation adherence as separate evidence unless strict formatting is enabled for the feature.
This run's encrypted evidence expires at the date shown above, 14 days after the run was created. The verdict summary and savings estimate remain in experiment history after that run-specific evidence is deleted.
Review the case results before opening sensitive prompt and response evidence. Opening a case is audited.
| Case | Result | Format | Evaluation confidence | Attempts | Production source | Replayed source | Candidate | Replay delta | Status | Case details |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Equivalent | Met | 97% | 2 | 1,170 ms 37,152 in / 1,742 out | 1,200 ms 37,152 in / 1,742 out | 510 ms 35,666 in / 1,463 out | -690 ms (-57%) | Evaluated |
Replay failures
0 source, 0 candidate
Failures are excluded from percentiles and retained as reliability evidence
| View details |
| 2 | Equivalent | Met | 97% | 1 | 2,182 ms 108,096 in / 5,111 out | 1,247 ms 108,096 in / 5,111 out | 541 ms 103,772 in / 4,293 out | -706 ms (-57%) | Evaluated | View details |
| 3 | Equivalent | Met | 97% | 1 | 478 ms 37,152 in / 1,921 out | 1,294 ms 37,152 in / 1,921 out | 572 ms 35,666 in / 1,614 out | -722 ms (-56%) | Evaluated | View details |
| 4 | Equivalent | Met | 97% | 1 | 3,542 ms 181,340 in / 8,935 out | 1,341 ms 181,340 in / 8,935 out | 603 ms 174,086 in / 7,505 out | -738 ms (-55%) | Evaluated | View details |
| 5 | Equivalent | Met | 97% | 1 | 797 ms 108,096 in / 4,530 out | 1,388 ms 108,096 in / 4,530 out | 634 ms 103,772 in / 3,805 out | -754 ms (-54%) | Evaluated | View details |
| 6 | Equivalent | Met | 97% | 1 | 1,399 ms 65,105 in / 2,573 out | 1,435 ms 65,105 in / 2,573 out | 665 ms 62,501 in / 2,161 out | -770 ms (-54%) | Evaluated | View details |
| 7 | Equivalent | Met | 97% | 1 | 1,119 ms 181,340 in / 9,829 out | 1,482 ms 181,340 in / 9,829 out | 696 ms 174,086 in / 8,256 out | -786 ms (-53%) | Evaluated | View details |
| 8 | Equivalent | Met | 97% | 1 | 569 ms 21,230 in / 1,099 out | 1,529 ms 21,230 in / 1,099 out | 727 ms 20,381 in / 923 out | -802 ms (-52%) | Evaluated | View details |
| 9 | Equivalent | Met | 97% | 1 | 361 ms 65,105 in / 2,859 out | 1,576 ms 65,105 in / 2,859 out | 758 ms 62,501 in / 2,402 out | -818 ms (-52%) | Evaluated | View details |
| 10 | Equivalent | Met | 97% | 1 | 108 ms 21,230 in / 965 out | 1,623 ms 21,230 in / 965 out | 529 ms 20,381 in / 811 out | -1,094 ms (-67%) | Evaluated | Hide details |
Case 10 evidenceFull captured evidence for this scored comparison. EquivalentFormat metscored97% evaluation confidenceJudge: gemini-3.1-pro-preview Evaluation notesThe candidate preserved the original answer's policy, resolution, and customer-safe guidance. Format evidenceThe response preserved the expected support-answer structure. Captured prompt | ||||||||||
Northstar support case 10: resolve the ticket-summary request with the account policy and cited help-center context.
claude-sonnet-4-6
Resolved ticket-summary case 10 with the correct policy, next step, and customer-safe explanation.
claude-haiku-4-5
Cheaper candidate preserved the policy, action, and explanation for ticket-summary case 10.