Anthropic
Claude Sonnet 5
Release date: 6/30/2026
A balanced frontier model designed for high-volume professional deployments.
Pramaana IndexAccuracy: StrongLatency: StrongCost: Strong
- Accuracy / Pramaana Index
- 68.61%± 0.99
- Latency / Pramaana Index
- 424.52s
- Cost / Test / Pramaana Index
- $1.85
- Context window
- 500k
- Max output tokens
- 96k
- Input modality
- Text · Image · File
Hyperparameter settings
temperature: 0.0
reasoning: extended
fallbacks: disabled
Benchmark performance for Claude Sonnet 5
| Benchmarks | Accuracy | Ranking |
|---|---|---|
| Pramaana Index | 68.61% ± 0.99 | 4 / 35 |
| Pramaana Multimodal Index | 64.53% ± 1.16 | 6 / 24 |
| Legal Research Fidelity | 66.41% ± 1.33 | 3 / 18 |
| Financial Modeling Agent | 58.72% ± 1.50 | 5 / 22 |
| Clinical Documentation Safety | 62.31% ± 1.67 | 4 / 19 |
| MATH-500 | 76.41% ± 0.48 | 6 / 112 |
| GPQA Diamond | 80.42% ± 0.65 | 6 / 121 |
| MMLU-Pro | 77.91% ± 0.82 | 5 / 119 |
| Adaptive Tutoring Dialogue | 64.21% ± 0.99 | 5 / 16 |
| SWE-bench Verified | 59.62% ± 1.16 | 4 / 69 |
| LiveCodeBench | 76.53% ± 1.33 | 4 / 127 |
| Agentic Reliability | 52.73% ± 1.50 | 6 / 12 |
| Strategic Games Arena | 60.82% ± 1.67 | 5 / 20 |
| Benefits Navigator | 64.12% ± 0.48 | 5 / 16 |