Gemini 4 Argon vs Gemini 4 Carbon

Every finance benchmark where both have a score (Google and Google). Data refreshed 2 Nov 2026.

Shared benchmarks
5
Gemini 4 Argon only
0
Gemini 4 Carbon only
0

Shared benchmarks, side by side

Each benchmark is drawn on its own scale. Bars are not comparable across rows.

  • Gemini 4 Argon
  • Gemini 4 Carbon
  • APEX-Accounting Mean Score (% criteria passed), higher is better

    Gemini 4 Argon:71.5 (better)
    Gemini 4 Carbon:70.7
  • Excel Modeling Benchmark (EMB) accuracy %, higher is better

    Gemini 4 Argon:74.8
    Gemini 4 Carbon:76.5 (better)
  • Finance Agent Benchmark Partial Credit accuracy %, higher is better

    Gemini 4 Argon:67.7 (better)
    Gemini 4 Carbon:67.2
  • Vals Finance Index index accuracy %, higher is better

    Gemini 4 Argon:64.9
    Gemini 4 Carbon:66.3 (better)
  • Tax Agent Bench accuracy % (all-pass x citation factor), higher is better

    Gemini 4 Argon:44.6
    Gemini 4 Carbon:48.9 (better)
Benchmark
Gemini 4 Argon
Gemini 4 Carbon

5 benchmarks, 5 scored for both. Bold marks the better score where both have one.

APEX-AccountingMean Score (% criteria passed), higher is better
71.5
Mercor APEX-Accounting leaderboard (Official board, 1 Nov 2026)
70.7
Mercor APEX-Accounting leaderboard (Official board, 1 Nov 2026)
Excel Modeling Benchmark (EMB)accuracy %, higher is better
74.8
Vals.ai leaderboard (Official board, 29 Oct 2026)
76.5
Vals.ai leaderboard (Official board, 29 Oct 2026)
Finance Agent BenchmarkPartial Credit accuracy %, higher is better
67.7
Vals.ai leaderboard (Official board, 29 Oct 2026)
67.2
Vals.ai leaderboard (Official board, 29 Oct 2026)
Vals Finance Indexindex accuracy %, higher is better
64.9
Vals.ai leaderboard (Official board, 29 Oct 2026)
66.3
Vals.ai leaderboard (Official board, 29 Oct 2026)
Tax Agent Benchaccuracy % (all-pass x citation factor), higher is better
44.6
Vals.ai leaderboard (Official board, 29 Oct 2026)
48.9
Vals.ai leaderboard (Official board, 29 Oct 2026)

Scores from different benchmarks use different scales and are not added up. Settings can differ between sources; see each leaderboard for details.