Model comparison
Selected 2 / 5See the differences. Choose what matters to you.
Available modelsSelected 2 / 5
Differences vs. k3.1-pro · max
Score, cost & time differences
Current baseline · k3.1-pro · max
84Backend engineering
Compared with k3.1-pro · max
- claude-opus-5-5 · max
Backend engineering: tiedReference cost $1.05 ($0.83 less)Evaluation time 9m 29s (33m 5s less)
Differences reflect these evaluation tasks. Cost and time cover the same dimension, not a typical conversation. Scoring & cost notes
| Metric | k3.1-pro · max | claude-opus-5-5 · max |
|---|---|---|
| Backend engineering | k3.1-promax 84 | claude-opus-5-5max 84Tied |
| Overall | k3.1-promax 79.60 | claude-opus-5-5max 89.15+9.55 pts |
| Frontend interaction | k3.1-promax 91 | claude-opus-5-5max 97+6 pts |
| Knowledge & Reasoning | k3.1-promax 80 | claude-opus-5-5max 100+20 pts |
| Visual aesthetics | k3.1-promax 68 | claude-opus-5-5max 73+5 pts |
| Maintainability | k3.1-promax 64 | claude-opus-5-5max 70+6 pts |
Backend engineering evaluation · Evaluation consumption, not a monthly subscription fee
Cost & time
Backend engineering evaluation| Metric | k3.1-pro · max | claude-opus-5-5 · max |
|---|---|---|
| Reference cost | k3.1-promax $1.87 | claude-opus-5-5max $1.05−$0.83 |
| Evaluation time | k3.1-promax 42m 34s | claude-opus-5-5max 9m 29s−77.7% |
Capability profiles
Show up to five models. Select names to focus the chart.
Behind each score
Expand to inspect each subaxis.
Backend engineeringSubscores
| Scoring dimension | k3.1-promax | claude-opus-5-5max |
|---|---|---|
| Contracts and boundaries | 16 / 20 | 18 / 20 |
| Retries and failures | 18 / 20 | 16 / 20 |
| CI and adversarial review | 13 / 20 | 16 / 20 |
| Transactions and consistency | 19 / 20 | 16 / 20 |
| Caching and regression | 18 / 20 | 18 / 20 |
Frontend interactionSubscores
| Scoring dimension | k3.1-promax | claude-opus-5-5max |
|---|---|---|
| Layout and adaptation | 20 / 20 | 20 / 20 |
| Interaction behavior | 20 / 20 | 20 / 20 |
| State continuity | 15 / 20 | 20 / 20 |
| Accessibility and feedback | 15 / 15 | 12 / 15 |
| State validation and regression | 21 / 25 | 25 / 25 |
Knowledge & ReasoningSubscores
| Scoring dimension | k3.1-promax | claude-opus-5-5max |
|---|---|---|
| Mathematical reasoning | 15 / 15 | 15 / 15 |
| Computer science & AI | 5 / 15 | 15 / 15 |
| Science and engineering | 10 / 10 | 10 / 10 |
| Life sciences and medicine | 15 / 20 | 20 / 20 |
| Humanities and general knowledge | 35 / 40 | 40 / 40 |
Visual aestheticsSubscores
| Scoring dimension | k3.1-promax | claude-opus-5-5max |
|---|---|---|
| Composition | 20 / 20 | 19 / 20 |
| Spacing and alignment | 15 / 20 | 18 / 20 |
| Typography and expression | 16 / 20 | 18 / 20 |
| Color and surfaces | 20 / 20 | 19 / 20 |
| Control styling | 17 / 20 | 16 / 20 |
Subscores retain their original scale; the axis score uses a consistent conversion.
MaintainabilitySubscores
| Scoring dimension | k3.1-promax | claude-opus-5-5max |
|---|---|---|
| Semantic clarity | 16 / 20 | 16 / 20 |
| Control-flow organization | 20 / 20 | 20 / 20 |
| Locality of changes | 15 / 20 | 18 / 20 |
| State and lifecycle | 18 / 20 | 18 / 20 |
| Implementation economy | 17 / 20 | 17 / 20 |
Subscores retain their original scale; the axis score uses a consistent conversion.