Agnes 2.5 Pro Beta vs GPT-6 Sol
Agnes 2.5 Pro Beta 
GPT-6 Sol Verdict
- GPT-6 Sol (max) leads on quality: 67.7 vs 60.0.
- Agnes 2.5 Pro Beta is stronger in agents & tools.
- GPT-6 Sol (max) is stronger in composite, knowledge, long context, multimodal, reasoning.
- Agnes 2.5 Pro Beta is 27× cheaper ($0.150 vs $4.00 per 1M blended).
| Metric | Agnes 2.5 Pro Beta | GPT-6 Sol (max) |
|---|---|---|
| BenchLeader Index | 60.0 | 67.7 |
| Agents & tools score | 64.4 | – |
| Composite score | 71.5 | 86.2 |
| Knowledge score | 58.0 | 75.4 |
| Long context score | 67.3 | 67.7 |
| Multimodal score | 59.0 | 67.1 |
| Reasoning score | 69.5 | 95.0 |
| Blended price $/M | $0.150 | $4.00 |
| Output speed | – | 126 tok/s |
| Time to first answer | – | 107.2 s |
| Context window | 1M | 1.1M |
| AA Intelligence Index | 35.2 | 47.5 |
| AA-LCR | 83.0% | 83.7% |
| MMMU-Pro | 75.4% | 83.3% |
| AA-Omniscience | -10.5 | 27.1 |
| GPQA Diamond (AA) | 90.5% | – |
| Humanity's Last Exam (AA) | 37.5% | 47.9% |
| SciCode (AA) | 48.8% | 57.6% |
| CritPt | 15.7% | 30.9% |
| GDPval (AA) | 40.6% | 49.4% |
| τ²-Bench Banking (AA) | 35.7% | – |
| Terminal-Bench 4.0 (AA) | – | 43.9% |
| Terminal-Bench 2.1 (AA) | 69.7% | – |
| AutomationBench | – | 61.6% |
| GDP.pdf | – | 24.8% |
| MLCR | – | 16.1% |
| AA-Omniscience: accuracy | 16.8% | 54.5% |
| AA-Omniscience: non-hallucination | 67.2% | 39.9% |
| AA-Briefcase | – | 1483 |
Data as of 2026-09-23. Best configuration of each model; every score links to its source on the model pages.
Agnes 2.5 Pro Beta vs GPT-6 Sol: questions
- Is Agnes 2.5 Pro Beta better than GPT-6 Sol?
- GPT-6 Sol (max) leads on quality: 67.7 vs 60.0. The BenchLeader Index combines every independent quality benchmark; GPT-6 Sol (max) is ahead overall as of 2026-09-23, but check the category scores for your use.
- Which is cheaper, Agnes 2.5 Pro Beta or GPT-6 Sol?
- Agnes 2.5 Pro Beta is cheaper: $0.150 against $4.00 per million tokens, blended at three input tokens per output token.
- Which has the larger context window?
- GPT-6 Sol accepts more context: 1.1M against 1M tokens.