# Claude Sonnet 5.5 vs Gemini 4 Argon

Every finance benchmark where both have a score (Anthropic and Google). Data refreshed 2 Nov 2026. Scores from different benchmarks use different scales and are not added up.

| Benchmark | Metric | Claude Sonnet 5.5 | Claude Sonnet 5.5 source | Gemini 4 Argon | Gemini 4 Argon source | Ahead |
| --- | --- | --- | --- | --- | --- | --- |
| Excel Modeling Benchmark (EMB) | accuracy % (v1.1) (higher is better) | 75.9 | Vals.ai leaderboard, 29 Oct 2026, https://www.vals.ai/benchmarks/emb-v1-1 | 74.8 | Vals.ai leaderboard, 29 Oct 2026, https://www.vals.ai/benchmarks/emb-v1-1 | Claude Sonnet 5.5 |
| Finance Agent Benchmark | Partial Credit accuracy % (v2.1) (higher is better) | 67.3 | Vals.ai leaderboard, 29 Oct 2026, https://www.vals.ai/benchmarks/fabv2-1 | 67.7 | Vals.ai leaderboard, 29 Oct 2026, https://www.vals.ai/benchmarks/fabv2-1 | Gemini 4 Argon |
| Vals Finance Index | index accuracy % (higher is better) | 64.5 | Vals.ai leaderboard, 29 Oct 2026, https://www.vals.ai/benchmarks/finance_index | 64.9 | Vals.ai leaderboard, 29 Oct 2026, https://www.vals.ai/benchmarks/finance_index | Gemini 4 Argon |
| Tax Agent Bench | accuracy % (all-pass x citation factor) (higher is better) | 42.0 | Vals.ai leaderboard, 29 Oct 2026, https://www.vals.ai/benchmarks/tax_agent_bench | 44.6 | Vals.ai leaderboard, 29 Oct 2026, https://www.vals.ai/benchmarks/tax_agent_bench | Gemini 4 Argon |

Source: https://benchmarks.finance/compare/claude-sonnet-5-5-vs-gemini-4-argon/

Created and maintained by Moneda.
