AI BENCHY Compare
Anthropic: Claude Sonnet 4.6 vs OpenAI: GPT-5.2 Chat
Vergelijken:
Benchmarks gegenereerd uit AI BENCHY-testsuites op: 2026-03-03
| Metriek | Anthropic: Claude Sonnet 4.6 medium Releasedatum: 2026-02-17 | OpenAI: GPT-5.2 Chat none Releasedatum: 2025-12-11 |
|---|---|---|
| Rang | #11 | #12 |
| Gem. score | 7.43 | 7.41 |
| Consistentie | 9.40 | 9.45 |
| Kosten per resultaat | 8.105 | 2.261 |
| Totale kosten | $0.811 | $0.227 |
| Correcte tests | ||
| Slaagpercentage per poging | 73.8% | 73.8% |
| Instabiele tests | 1 | 1 |
| Uitvoer-tokens | 29,098 | 14,267 |
| Redeneer-tokens | 20,435 | 0 |
Score vs totale kosten
Categorie-uitsplitsing
| Anti-AI-trucs | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 7.00 | 10.00 | 66.7% | 0 | 1,031 | 1,093 | |
| OpenAI: GPT-5.2 Chat | 10.00 | 10.00 | 100.0% | 0 | 1,651 | 0 |
| Gegevensparsering en extractie | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 10.00 | 10.00 | 100.0% | 0 | 727 | 907 | |
| OpenAI: GPT-5.2 Chat | 9.88 | 10.00 | 100.0% | 0 | 980 | 0 |
| Domeinspecifiek | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 1.00 | 7.21 | 11.1% | 1 | 25,790 | 16,919 | |
| OpenAI: GPT-5.2 Chat | 4.00 | 10.00 | 33.3% | 0 | 7,810 | 0 |
| Instructies opvolgen | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 10.00 | 10.00 | 100.0% | 0 | 316 | 523 | |
| OpenAI: GPT-5.2 Chat | 5.50 | 6.13 | 66.7% | 1 | 1,528 | 0 |
| Puzzle Solving | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 10.00 | 10.00 | 100.0% | 0 | 579 | 642 | |
| OpenAI: GPT-5.2 Chat | 7.00 | 10.00 | 66.7% | 0 | 1,743 | 0 |
| Toolaanroepen | Score | Consistentie | Slaagpercentage per poging | Instabiele tests | Correcte tests | Uitvoer-tokens | Redeneer-tokens |
|---|---|---|---|---|---|---|---|
| Anthropic: Claude Sonnet 4.6 | 10.00 | 10.00 | 100.0% | 0 | 655 | 351 | |
| OpenAI: GPT-5.2 Chat | 10.00 | 10.00 | 100.0% | 0 | 555 | 0 |
Snelle vergelijking
Vergelijkingspaar wisselen
GPT-5.2 ChatnonevsGLM 5mediumClaude Sonnet 4.6mediumvsGPT-5.3 ChatnoneClaude Sonnet 4.6mediumvsGemini 3 Flash PreviewnoneClaude Sonnet 4.6mediumvsGemini 3.1 Flash Lite PreviewnoneClaude Sonnet 4.6mediumvsGemini 3.1 Flash Lite PreviewhighGemini 2.5 FlashmediumvsGPT-5.2 ChatnoneGemini 3.1 Flash Lite PreviewhighvsGPT-5.2 ChatnoneClaude Sonnet 4.6mediumvsGemini 3.1 Flash Lite PreviewlowGPT-5.2 ChatnonevsStep 3.5 FlashmediumGratis beschikbaarGemini 3.1 Flash Lite PreviewlowvsGPT-5.2 ChatnoneDeepSeek V3.2mediumvsGPT-5.2 ChatnoneGemini 3.1 Flash Lite PreviewmediumvsGPT-5.2 Chatnone