Claude Sonnet 5.5 vs Grok 4.7
Grok 4.7 is 1.3× cheaper: $3.00 per million tokens blended, against $4.00 for Claude Sonnet 5.5. On the Vals Index v2.2, Claude Sonnet 5.5 scores 70.85% and Grok 4.7 58.62%, so Claude Sonnet 5.5 leads.
Price
| Per million tokens | Claude Sonnet 5.5 | Grok 4.7 |
|---|---|---|
| Maker | Anthropic | xAI |
| Input | $2.00 | $2.00 |
| Output | $10.00 | $6.00 |
| Cache read | $0.1 | $0.5 |
| Blended (3:1) | $4.00 | $3.00 |
| Context window | n/a | 500K |
| Output speed | 135 tokens/s | 94 tokens/s |
| Status | Available | Available |
Sources
Anthropic: official API pricing, read 2026-11-12
xAI: Grok model pricing, read 2026-11-12
Benchmarks
| Benchmark | Claude Sonnet 5.5 | Grok 4.7 |
|---|---|---|
| Vals Index v2.2 Vals AI | 70.85% | 58.62% |
| Artificial Analysis Intelligence Index Artificial Analysis | 55max | 35xhigh |
| Epoch Capabilities Index Epoch AI | 166.6 | 156.4 |
| Arena Text Arena | 1473xHigh | 1443xHigh |
| Terminal-Bench 4.0 Vals AI | 64.14% | 28.79% |