Model comparison
Selected 2 / 5See the differences. Choose what matters to you.
Available modelsSelected 2 / 5
Differences vs. DeepSeek V4 Flash · high
Score, cost & time differences
Current baseline · DeepSeek V4 Flash · high
58.00Overall
Compared with DeepSeek V4 Flash · high
- DeepSeek V4 Flash · max
Overall: −3.4 ptsCost $0.60 ($0.18 more)Evaluation time 89m 12s (9m 13s less)
Differences reflect these evaluation tasks. Cost and time cover the same dimension, not a typical conversation. Scoring & cost notes
| Metric | DeepSeek V4 Flash · high | DeepSeek V4 Flash · max |
|---|---|---|
| Overall | DeepSeek V4 Flashhigh 58.00 | DeepSeek V4 Flashmax 54.60−3.4 pts |
| Backend engineering | DeepSeek V4 Flashhigh 53 | DeepSeek V4 Flashmax 54+1 pts |
| Frontend interaction | DeepSeek V4 Flashhigh 67 | DeepSeek V4 Flashmax 55−12 pts |
| Knowledge & reasoning | DeepSeek V4 Flashhigh 55 | DeepSeek V4 Flashmax 55Tied |
| Visual aesthetics | DeepSeek V4 Flashhigh 61 | DeepSeek V4 Flashmax 61Tied |
| Maintainability | DeepSeek V4 Flashhigh 53 | DeepSeek V4 Flashmax 47−6 pts |
Complete evaluation · Evaluation consumption, not a monthly subscription fee
Cost & time
Complete evaluation| Metric | DeepSeek V4 Flash · high | DeepSeek V4 Flash · max |
|---|---|---|
| Cost | DeepSeek V4 Flashhigh $0.42 | DeepSeek V4 Flashmax $0.60+$0.18 |
| Evaluation time | DeepSeek V4 Flashhigh 98m 25s | DeepSeek V4 Flashmax 89m 12s−9.4% |
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 | DeepSeek V4 Flashhigh | DeepSeek V4 Flashmax |
|---|---|---|
| Contracts and boundaries | 8 / 20 | 10 / 20 |
| Retries and failures | 16 / 20 | 17 / 20 |
| CI and adversarial review | 11 / 20 | 12 / 20 |
| Transactions and consistency | 10 / 20 | 15 / 20 |
| Caching and regression | 8 / 20 | 0 / 20 |
Frontend interactionSubscores
| Scoring dimension | DeepSeek V4 Flashhigh | DeepSeek V4 Flashmax |
|---|---|---|
| Layout and adaptation | 16 / 20 | 10 / 20 |
| Interaction behavior | 16 / 20 | 9 / 20 |
| State continuity | 12 / 20 | 6 / 20 |
| Accessibility and feedback | 8 / 15 | 8 / 15 |
| State validation and regression | 15 / 25 | 22 / 25 |
Knowledge & reasoningSubscores
| Scoring dimension | DeepSeek V4 Flashhigh | DeepSeek V4 Flashmax |
|---|---|---|
| Mathematical reasoning | 10 / 15 | 5 / 15 |
| Computer science & AI | 10 / 15 | 10 / 15 |
| Science and engineering | 5 / 10 | 0 / 10 |
| Life sciences and medicine | 15 / 20 | 15 / 20 |
| Humanities and general knowledge | 15 / 40 | 25 / 40 |
Visual aestheticsSubscores
| Scoring dimension | DeepSeek V4 Flashhigh | DeepSeek V4 Flashmax |
|---|---|---|
| Composition | 19 / 20 | 19 / 20 |
| Spacing and alignment | 15 / 20 | 16 / 20 |
| Typography and expression | 16 / 20 | 17 / 20 |
| Color and surfaces | 19 / 20 | 19 / 20 |
| Control styling | 16 / 20 | 14 / 20 |
Subscores retain their original scale; the axis score uses a consistent conversion.
MaintainabilitySubscores
| Scoring dimension | DeepSeek V4 Flashhigh | DeepSeek V4 Flashmax |
|---|---|---|
| Semantic clarity | 15 / 20 | 15 / 20 |
| Control-flow organization | 19 / 20 | 16 / 20 |
| Locality of changes | 15 / 20 | 15 / 20 |
| State and lifecycle | 16 / 20 | 16 / 20 |
| Implementation economy | 16 / 20 | 16 / 20 |
Subscores retain their original scale; the axis score uses a consistent conversion.