Model comparison
Selected 2 / 5See the differences. Choose what matters to you.
Differences vs. claude-haiku-5-5 · xhigh
Score, cost & time differences
Current baseline · claude-haiku-5-5 · xhigh
60Knowledge & Reasoning
Compared with claude-haiku-5-5 · xhigh
- deepseek-v4-flash · high
Knowledge & Reasoning: −5 ptsReference cost $0.26 ($0.19 more)Evaluation time 70m 1s (58m 34s more)
Differences reflect these evaluation tasks. Cost and time cover the same dimension, not a typical conversation. Scoring & cost notes
| Metric | claude-haiku-5-5 · xhigh | deepseek-v4-flash · high |
|---|---|---|
| Knowledge & Reasoning | claude-haiku-5-5xhigh 60 | deepseek-v4-flashhigh 55−5 pts |
| Overall | claude-haiku-5-5xhigh 65.65 | deepseek-v4-flashhigh 60.50−5.15 pts |
| Backend engineering | claude-haiku-5-5xhigh 69 | deepseek-v4-flashhigh 58−11 pts |
| Frontend interaction | claude-haiku-5-5xhigh 78 | deepseek-v4-flashhigh 76−2 pts |
| Visual aesthetics | claude-haiku-5-5xhigh 59 | deepseek-v4-flashhigh 61+2 pts |
| Maintainability | claude-haiku-5-5xhigh 64 | deepseek-v4-flashhigh 53−11 pts |
Knowledge & Reasoning evaluation · Evaluation consumption, not a monthly subscription fee
Cost & time
Knowledge & Reasoning evaluation| Metric | claude-haiku-5-5 · xhigh | deepseek-v4-flash · high |
|---|---|---|
| Reference cost | claude-haiku-5-5xhigh $0.068 | deepseek-v4-flashhigh $0.26+$0.19 |
| Evaluation time | claude-haiku-5-5xhigh 11m 27s | deepseek-v4-flashhigh 70m 1s+511.7% |
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 | claude-haiku-5-5xhigh | deepseek-v4-flashhigh |
|---|---|---|
| Contracts and boundaries | 15 / 20 | 8 / 20 |
| Retries and failures | 15 / 20 | 16 / 20 |
| CI and adversarial review | 12 / 20 | 11 / 20 |
| Transactions and consistency | 12 / 20 | 10 / 20 |
| Caching and regression | 15 / 20 | 13 / 20 |
Frontend interactionSubscores
| Scoring dimension | claude-haiku-5-5xhigh | deepseek-v4-flashhigh |
|---|---|---|
| Layout and adaptation | 18 / 20 | 16 / 20 |
| Interaction behavior | 20 / 20 | 20 / 20 |
| State continuity | 15 / 20 | 15 / 20 |
| Accessibility and feedback | 10 / 15 | 10 / 15 |
| State validation and regression | 15 / 25 | 15 / 25 |
Knowledge & ReasoningSubscores
| Scoring dimension | claude-haiku-5-5xhigh | deepseek-v4-flashhigh |
|---|---|---|
| Mathematical reasoning | 10 / 15 | 10 / 15 |
| Computer science & AI | 5 / 15 | 10 / 15 |
| Science and engineering | 5 / 10 | 5 / 10 |
| Life sciences and medicine | 15 / 20 | 15 / 20 |
| Humanities and general knowledge | 25 / 40 | 15 / 40 |
Visual aestheticsSubscores
| Scoring dimension | claude-haiku-5-5xhigh | deepseek-v4-flashhigh |
|---|---|---|
| Composition | 19 / 20 | 19 / 20 |
| Spacing and alignment | 16 / 20 | 15 / 20 |
| Typography and expression | 16 / 20 | 16 / 20 |
| Color and surfaces | 18 / 20 | 19 / 20 |
| Control styling | 15 / 20 | 16 / 20 |
Subscores retain their original scale; the axis score uses a consistent conversion.
MaintainabilitySubscores
| Scoring dimension | claude-haiku-5-5xhigh | deepseek-v4-flashhigh |
|---|---|---|
| Semantic clarity | 15 / 20 | 15 / 20 |
| Control-flow organization | 20 / 20 | 19 / 20 |
| Locality of changes | 17 / 20 | 15 / 20 |
| State and lifecycle | 18 / 20 | 16 / 20 |
| Implementation economy | 16 / 20 | 16 / 20 |
Subscores retain their original scale; the axis score uses a consistent conversion.