Model comparison
Selected 2 / 5See the differences. Choose what matters to you.
Differences vs. ERNIE 5.5 Flash Preview · Max
Score, cost & time differences
Current baseline · ERNIE 5.5 Flash Preview · Max
66Visual aesthetics
Compared with ERNIE 5.5 Flash Preview · Max
- DeepSeek V4.1 Flash (1017) · high
Visual aesthetics: tiedComparable cost unavailableComparable time unavailable
Differences reflect these evaluation tasks. Cost and time cover the same dimension, not a typical conversation. Scoring & cost notes
| Metric | ERNIE 5.5 Flash Preview · Max | DeepSeek V4.1 Flash (1017) · high |
|---|---|---|
| Visual aesthetics | ERNIE 5.5 Flash PreviewMax 66 | DeepSeek V4.1 Flash (1017)high 66Tied |
| Overall | ERNIE 5.5 Flash PreviewMax 67.20 | DeepSeek V4.1 Flash (1017)high 69.25+2.05 pts |
| Backend engineering | ERNIE 5.5 Flash PreviewMax 52 | DeepSeek V4.1 Flash (1017)high 60+8 pts |
| Frontend interaction | ERNIE 5.5 Flash PreviewMax 52 | DeepSeek V4.1 Flash (1017)high 81+29 pts |
| Knowledge & reasoning | ERNIE 5.5 Flash PreviewMax 90 | DeepSeek V4.1 Flash (1017)high 70−20 pts |
| Maintainability | ERNIE 5.5 Flash PreviewMax 57 | DeepSeek V4.1 Flash (1017)high 59+2 pts |
Visual aesthetics evaluation · Evaluation consumption, not a monthly subscription fee
Cost & time
Visual aesthetics evaluation| Metric | ERNIE 5.5 Flash Preview · Max | DeepSeek V4.1 Flash (1017) · high |
|---|---|---|
| Cost | ERNIE 5.5 Flash PreviewMax N/A | DeepSeek V4.1 Flash (1017)high N/A |
| Evaluation time | ERNIE 5.5 Flash PreviewMax — | DeepSeek V4.1 Flash (1017)high — |
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 | ERNIE 5.5 Flash PreviewMax | DeepSeek V4.1 Flash (1017)high |
|---|---|---|
| Contracts and boundaries | 7 / 20 | 7 / 20 |
| Retries and failures | 10 / 20 | 17 / 20 |
| CI and adversarial review | 14 / 20 | 12 / 20 |
| Transactions and consistency | 12 / 20 | 13 / 20 |
| Caching and regression | 9 / 20 | 11 / 20 |
Frontend interactionSubscores
| Scoring dimension | ERNIE 5.5 Flash PreviewMax | DeepSeek V4.1 Flash (1017)high |
|---|---|---|
| Layout and adaptation | — | 18 / 20 |
| Interaction behavior | — | 18 / 20 |
| State continuity | — | 11 / 20 |
| Accessibility and feedback | 5 / 15 | 13 / 15 |
| State validation and regression | — | 21 / 25 |
Knowledge & reasoningSubscores
| Scoring dimension | ERNIE 5.5 Flash PreviewMax | DeepSeek V4.1 Flash (1017)high |
|---|---|---|
| Mathematical reasoning | 15 / 15 | 15 / 15 |
| Computer science & AI | 10 / 15 | 10 / 15 |
| Science and engineering | 10 / 10 | 5 / 10 |
| Life sciences and medicine | 20 / 20 | 20 / 20 |
| Humanities and general knowledge | 35 / 40 | 20 / 40 |
Visual aestheticsSubscores
| Scoring dimension | ERNIE 5.5 Flash PreviewMax | DeepSeek V4.1 Flash (1017)high |
|---|---|---|
| Composition | 18 / 20 | 19 / 20 |
| Spacing and alignment | 18 / 20 | 16 / 20 |
| Typography and expression | 16 / 20 | 16 / 20 |
| Color and surfaces | 19 / 20 | 19 / 20 |
| Control styling | 16 / 20 | 17 / 20 |
Subscores retain their original scale; the axis score uses a consistent conversion.
MaintainabilitySubscores
| Scoring dimension | ERNIE 5.5 Flash PreviewMax | DeepSeek V4.1 Flash (1017)high |
|---|---|---|
| Semantic clarity | 15 / 20 | 16 / 20 |
| Control-flow organization | 19 / 20 | 19 / 20 |
| Locality of changes | 15 / 20 | 15 / 20 |
| State and lifecycle | 17 / 20 | 17 / 20 |
| Implementation economy | 17 / 20 | 17 / 20 |
Subscores retain their original scale; the axis score uses a consistent conversion.