Category scores
Out of 100: 100 = the best tracked model in the category.
| Reasoning | 96 | |
| Mathematics | 99 | |
| Code | 100 | |
| Agents & real work | 80 | |
| Knowledge | 96 | |
| Long context | 91 | |
| Multimodal | 99 |
EditionUpdated daily
Ranking / GPT-6 Astra
GPT-6 Astra
OpenAI · Proprietary · context 1.05M. Ranked #1 of 26 with 929 ERAZ points out of 1000.
ERAZ rank#1/ 26
Points929/ 1000
Coverage95%
Benchmarks won6/ 14
Input price / M$10
Output price / M$50
Points over time
| Day | Points | Change |
|---|
Results on each benchmark
| Benchmark | Score | Rank | Out of 100 | Source |
|---|---|---|---|---|
| Humanity's Last Exam | 54.7% | #6 / 26 | 89 | src ↗ |
| ARC-AGI-2 | 95% | #1 / 13 | 100 | src ↗ |
| ARC-AGI-3 | 62.7% | #1 / 9 | 100 | src ↗ |
| FrontierMath (Tiers 1-3) | 93.7% | #1 / 16 | 100 | src ↗ |
| FrontierMath (Tier 4) | 97.6% | #2 / 16 | 98 | src ↗ |
| SWE-bench Pro | not published | — | — | |
| FrontierSWE v2 | 65.5% | #1 / 14 | 100 | src ↗ |
| Terminal-Bench 2.1 | 87.3% | #1 / 12 | 100 | src ↗ |
| OSWorld 2.0 | 72.6% | #1 / 11 | 100 | src ↗ |
| GDPval-AA | 1573 | #18 / 23 | 66 | src ↗ |
| GPQA Diamond | 96.1% | #2 / 15 | 99 | src ↗ |
| AA-Omniscience | 43.4 | #3 / 26 | 94 | src ↗ |
| AA-LCR | 80.7% | #16 / 26 | 91 | src ↗ |
| MMMU-Pro | 86.9% | #2 / 14 | 99 | src ↗ |
Artificial Analysis Intelligence Index : 52.7 · Source page ↗
Nearby models in the ranking
</> Embed this ranking
Show GPT-6 Astra's ranking on your site or in a README. The badge updates every day.
