Model comparison
Selected 2 / 5See the differences. Choose what matters to you.
Available modelsSelected 2 / 5
Differences vs. gpt-5.6-terra · max
Score, cost & time differences
Current baseline · gpt-5.6-terra · max
65.90Overall
Compared with gpt-5.6-terra · max
- claude-haiku-5-5 · xhigh
Overall: −0.25 ptsReference cost $0.21 ($1.73 less)Evaluation time 27m 38s (105m 24s less)
Differences reflect these evaluation tasks. Cost and time cover the same dimension, not a typical conversation. Scoring & cost notes
| Metric | gpt-5.6-terra · max | claude-haiku-5-5 · xhigh |
|---|---|---|
| Overall | gpt-5.6-terramax 65.90 | claude-haiku-5-5xhigh 65.65−0.25 pts |
| Backend engineering | gpt-5.6-terramax 81 | claude-haiku-5-5xhigh 69−12 pts |
| Frontend interaction | gpt-5.6-terramax 76 | claude-haiku-5-5xhigh 78+2 pts |
| Knowledge & Reasoning | gpt-5.6-terramax 55 | claude-haiku-5-5xhigh 60+5 pts |
| Visual aesthetics | gpt-5.6-terramax 61 | claude-haiku-5-5xhigh 59−2 pts |
| Maintainability | gpt-5.6-terramax 61 | claude-haiku-5-5xhigh 64+3 pts |
Complete evaluation · Evaluation consumption, not a monthly subscription fee
Cost & time
Complete evaluation| Metric | gpt-5.6-terra · max | claude-haiku-5-5 · xhigh |
|---|---|---|
| Reference cost | gpt-5.6-terramax $1.94 | claude-haiku-5-5xhigh $0.21−$1.73 |
| Evaluation time | gpt-5.6-terramax 133m 2s | claude-haiku-5-5xhigh 27m 38s−79.2% |
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 | gpt-5.6-terramax | claude-haiku-5-5xhigh |
|---|---|---|
| Contracts and boundaries | 17 / 20 | 15 / 20 |
| Retries and failures | 17 / 20 | 15 / 20 |
| CI and adversarial review | 12 / 20 | 12 / 20 |
| Transactions and consistency | 17 / 20 | 12 / 20 |
| Caching and regression | 18 / 20 | 15 / 20 |
Frontend interactionSubscores
| Scoring dimension | gpt-5.6-terramax | claude-haiku-5-5xhigh |
|---|---|---|
| Layout and adaptation | 18 / 20 | 18 / 20 |
| Interaction behavior | 17 / 20 | 20 / 20 |
| State continuity | 10 / 20 | 15 / 20 |
| Accessibility and feedback | 12 / 15 | 10 / 15 |
| State validation and regression | 19 / 25 | 15 / 25 |
Knowledge & ReasoningSubscores
| Scoring dimension | gpt-5.6-terramax | claude-haiku-5-5xhigh |
|---|---|---|
| Mathematical reasoning | 10 / 15 | 10 / 15 |
| Computer science & AI | 10 / 15 | 5 / 15 |
| Science and engineering | 0 / 10 | 5 / 10 |
| Life sciences and medicine | 15 / 20 | 15 / 20 |
| Humanities and general knowledge | 20 / 40 | 25 / 40 |
Visual aestheticsSubscores
| Scoring dimension | gpt-5.6-terramax | claude-haiku-5-5xhigh |
|---|---|---|
| Composition | 18 / 20 | 19 / 20 |
| Spacing and alignment | 16 / 20 | 16 / 20 |
| Typography and expression | 17 / 20 | 16 / 20 |
| Color and surfaces | 18 / 20 | 18 / 20 |
| Control styling | 16 / 20 | 15 / 20 |
Subscores retain their original scale; the axis score uses a consistent conversion.
MaintainabilitySubscores
| Scoring dimension | gpt-5.6-terramax | claude-haiku-5-5xhigh |
|---|---|---|
| Semantic clarity | 15 / 20 | 15 / 20 |
| Control-flow organization | 20 / 20 | 20 / 20 |
| Locality of changes | 16 / 20 | 17 / 20 |
| State and lifecycle | 18 / 20 | 18 / 20 |
| Implementation economy | 16 / 20 | 16 / 20 |
Subscores retain their original scale; the axis score uses a consistent conversion.