| Result | Rank 1GPT-6 Luna Low | Rank 2Claude Sonnet 5.5 Low | Rank 3Claude Opus 5.5 Low | Rank 4Grok 4.6 Low | Rank 5Kimi K3 Low |
|---|---|---|---|---|---|
| Combined score | 90.48 | 86.81 | 84.14 | 83.36 | 73.09 |
| Judge | |||||
| Judge score /100 | 90.00 | 90.00 | 87.50 | 86.00 | 74.00 |
| Correctness /10 | 9 | 9 | 9 | 9 | 7 |
| Coverage /10 | 9 | 9 | 8 | 8 | 8 |
| Evidence /10 | 9 | 9 | 9 | 9 | 7 |
| Actionability /10 | 9 | 9 | 9 | 8 | 8 |
| Time | 3.92 s | 9.63 s | 8.79 s | 18.18 s | 15.16 s |
| Estimated USD | $0.0004364 | $0.023284 | $0.038616 | $0.013568 | $0.0100031 |
Cost estimates are reference estimates, not billed charges.
Frozen reference rate sheet: openrouter-rate-sheet-v1:bb0fcbc9acd0e72b8495b144bad824b91029312178cb40b6e4e35c653df4a163. API-equivalent estimate; subscription quota and charges are unknown.
Judge score uses recorded rubric metrics; final rank uses the combined score.
Component points and token usage
| Result | Rank 1 GPT-6 Luna Low | Rank 2 Claude Sonnet 5.5 Low | Rank 3 Claude Opus 5.5 Low | Rank 4 Grok 4.6 Low | Rank 5 Kimi K3 Low |
|---|---|---|---|---|---|
| Quality points | 81 | 81 | 78.75 | 77.4 | 66.6 |
| Test points | 0 | 0 | 0 | 0 | 0 |
| Speed points | 4.6936 | 4.3088 | 4.3609 | 3.8373 | 3.9914 |
| Cost points | 4.7909 | 1.5022 | 1.0285 | 2.1215 | 2.4996 |
| Total input incl cache | 2,799 | 6,232 | 6,224 | 3,136 | 4,133 |
| Cache read (subset of input) | 0 | 2,238 | 2,234 | 512 | 0 |
| Cache write (subset of input) | 0 | 3,992 | 3,988 | 0 | 0 |
| Total output incl reasoning | 313 | 1,082 | 686 | 1,216 | 474 |
| Reasoning (subset of output) | N/A | N/A | N/A | N/A | N/A |
Swipe, scroll or use Previous and More to compare configurations. Row labels stay visible.
Advanced results and scoring details
ux · 2026-10-14T06:16:26.670Z · Prospective evaluation. Equal transport and native system conditions are unverified.
Applied weights: quality 90%, tests 0%, speed 5%, cost 5%.
Displayed scores are rounded; pairwise outcomes use exact recorded fractions. Reference costs are estimates, not billed charges. Missing components follow the recorded assessment.
Pairwise results
- Kimi K3 Low vs GPT-6 Luna Low: Second configuration wins
- Kimi K3 Low vs Grok 4.6 Low: Second configuration wins
- Kimi K3 Low vs Claude Sonnet 5.5 Low: Second configuration wins
- Kimi K3 Low vs Claude Opus 5.5 Low: Second configuration wins
- GPT-6 Luna Low vs Grok 4.6 Low: First configuration wins
- GPT-6 Luna Low vs Claude Sonnet 5.5 Low: First configuration wins
- GPT-6 Luna Low vs Claude Opus 5.5 Low: First configuration wins
- Grok 4.6 Low vs Claude Sonnet 5.5 Low: Second configuration wins
- Grok 4.6 Low vs Claude Opus 5.5 Low: Second configuration wins
- Claude Sonnet 5.5 Low vs Claude Opus 5.5 Low: First configuration wins