Model comparison
Selected 2 / 5See the differences. Choose what matters to you.
Available modelsSelected 2 / 5
Differences vs. HY4-flash · default
Score, cost & time differences
Current baseline · HY4-flash · default
42.2Overall
Compared with HY4-flash · default
- GPT-6 Luna · xhigh
Overall: tiedCost $0.87 ($0.39 more)Evaluation time 140m 31s (0m 57s more)
Differences reflect these evaluation tasks. Cost and time cover the same dimension, not a typical conversation. Scoring & cost notes
| Metric | HY4-flash · default | GPT-6 Luna · xhigh |
|---|---|---|
| Overall | HY4-flashdefault 42.2 | GPT-6 Lunaxhigh 42.2Tied |
| Backend engineering | HY4-flashdefault 38.8 | GPT-6 Lunaxhigh 35.7−3.1 pts |
| Visual aesthetics | HY4-flashdefault 25.0 | GPT-6 Lunaxhigh 21.0−4 pts |
| Knowledge & reasoning | HY4-flashdefault 55.0 | GPT-6 Lunaxhigh 50.0−5 pts |
| Frontend engineering | HY4-flashdefault 40.3 | GPT-6 Lunaxhigh 56.8+16.5 pts |
| Maintainability | HY4-flashdefault 47.0 | GPT-6 Lunaxhigh 49.0+2 pts |
Complete evaluation · Evaluation consumption, not a monthly subscription fee
Cost & time
Complete evaluation| Metric | HY4-flash · default | GPT-6 Luna · xhigh |
|---|---|---|
| Cost | HY4-flashdefault $0.47 | GPT-6 Lunaxhigh $0.87+$0.39 |
| Evaluation time | HY4-flashdefault 139m 34s | GPT-6 Lunaxhigh 140m 31s+0.7% |
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 | HY4-flashdefault | GPT-6 Lunaxhigh |
|---|---|---|
| Contracts and exceptional boundaries | 36.4 / 100.0 | 35.1 / 100.0 |
| State consistency and persistence | 58.7 / 100.0 | 53.1 / 100.0 |
| Dependencies, scheduling and resources | 79.0 / 100.0 | 62.9 / 100.0 |
| Queries and data transformations | 1.6 / 100.0 | 7.0 / 100.0 |
Visual aestheticsSubscores
| Scoring dimension | HY4-flashdefault | GPT-6 Lunaxhigh |
|---|---|---|
| Content hierarchy and readability | 13.0 / 20.0 | 13.0 / 20.0 |
| Spatial organization and information capacity | 15.0 / 20.0 | 15.0 / 20.0 |
| Responsive adaptation and layout | 15.0 / 20.0 | 13.0 / 20.0 |
| Control craft and visual distinction | 13.0 / 15.0 | 13.0 / 15.0 |
| Editing context and status expression | 7.0 / 25.0 | 5.0 / 25.0 |
Subscores retain their original scale; the axis score uses a consistent conversion.
Knowledge & reasoningSubscores
| Scoring dimension | HY4-flashdefault | GPT-6 Lunaxhigh |
|---|---|---|
| Mathematical reasoning | 15.0 / 15.0 | 0.0 / 15.0 |
| Computer science & AI | 10.0 / 15.0 | 5.0 / 15.0 |
| Science and engineering | 5.0 / 10.0 | 10.0 / 10.0 |
| Life sciences and medicine | 10.0 / 20.0 | 15.0 / 20.0 |
| Humanities and general knowledge | 15.0 / 40.0 | 20.0 / 40.0 |
Frontend engineeringSubscores
| Scoring dimension | HY4-flashdefault | GPT-6 Lunaxhigh |
|---|---|---|
| Environment adaptation | 5.2 / 10.0 | 5.2 / 10.0 |
| Interaction control | 12.0 / 17.0 | 11.3 / 17.0 |
| Intent preservation | 8.1 / 17.0 | 13.5 / 17.0 |
| State consistency | 4.6 / 17.0 | 10.4 / 17.0 |
| Fault recovery | 0.0 / 12.0 | 5.1 / 12.0 |
| Task continuity | 3.1 / 9.0 | 4.1 / 9.0 |
| Operability | 3.8 / 12.0 | 7.1 / 12.0 |
| Understandable feedback | 3.4 / 6.0 | 0.0 / 6.0 |
MaintainabilitySubscores
| Scoring dimension | HY4-flashdefault | GPT-6 Lunaxhigh |
|---|---|---|
| Code comprehension and responsibility | 23.0 / 25.0 | 22.0 / 25.0 |
| Change isolation and reuse | 21.0 / 25.0 | 22.0 / 25.0 |
| State and lifecycle | 16.0 / 20.0 | 16.0 / 20.0 |
| Complexity and necessary mechanisms | 11.0 / 20.0 | 12.0 / 20.0 |
| Diagnosis and behavioral verification | 6.5 / 10.0 | 7.0 / 10.0 |
Subscores retain their original scale; the axis score uses a consistent conversion.