RLTasksby ulam.ai
Models / Google

Google / MODEL EVIDENCE

Gemini 2.5 Flash Lite (9/25)

13 benchmarks · 28 observations · 14 sources

Scores at a glance

13 benchmarks · large dot: Gemini 2.5 Flash Lite (9/25); small dots: closest models where reported

LegalBench ↗79.01 %
79.0179.01

Gemini 2.5 Flash Lite (9/25) (Nonthinking) · LegalBench · Accuracy ↗

1 more reported result

82.04 % · Gemini 2.5 Flash Lite (9/25) (Thinking) · LegalBench · Accuracy · Vals AI · LegalBench ↗

CorpFin v2 ↗56.29 %
56.2956.29

Gemini 2.5 Flash Lite (9/25) (Nonthinking) · CorpFin · Accuracy ↗

1 more reported result

57.58 % · Gemini 2.5 Flash Lite (9/25) (Thinking) · CorpFin · Accuracy · Vals AI · CorpFin ↗

MedCode ↗27.08 %
27.0827.08

Gemini 2.5 Flash Lite (9/25) (Nonthinking) · MedCode · Accuracy ↗

1 more reported result

34.19 % · Gemini 2.5 Flash Lite (9/25) (Thinking) · MedCode · Accuracy · Vals AI · MedCode ↗

MedScribe ↗75.82 %
75.8275.82

Gemini 2.5 Flash Lite (9/25) (Nonthinking) · MedScribe · Accuracy ↗

1 more reported result

66.88 % · Gemini 2.5 Flash Lite (9/25) (Thinking) · MedScribe · Accuracy · Vals AI · MedScribe ↗

MedQA (USMLE) ↗80.33 %
80.3380.33

Gemini 2.5 Flash Lite (9/25) (Nonthinking) · MedQA · Accuracy ↗

1 more reported result

88.87 % · Gemini 2.5 Flash Lite (9/25) (Thinking) · MedQA · Accuracy · Vals AI · MedQA ↗

GPQA Diamond ↗64.14 %
64.1464.14

Gemini 2.5 Flash Lite (9/25) (Nonthinking) · GPQA Diamond · Accuracy ↗

1 more reported result

70.2 % · Gemini 2.5 Flash Lite (9/25) (Thinking) · GPQA Diamond · Accuracy · Vals AI · GPQA Diamond ↗

MMLU-Pro ↗78.64 %
78.6478.64

Gemini 2.5 Flash Lite (9/25) (Nonthinking) · MMLU Pro · Accuracy ↗

1 more reported result

79.12 % · Gemini 2.5 Flash Lite (9/25) (Thinking) · MMLU Pro · Accuracy · Vals AI · MMLU Pro ↗

MMMU-Pro ↗72.54 %
72.5472.54

Gemini 2.5 Flash Lite (9/25) (Nonthinking) · MMMU Pro · Accuracy ↗

1 more reported result

75.43 % · Gemini 2.5 Flash Lite (9/25) (Thinking) · MMMU Pro · Accuracy · Vals AI · MMMU Pro ↗

LiveCodeBench ↗67.67 %
67.6767.67

Gemini 2.5 Flash Lite (9/25) (Nonthinking) · LiveCodeBench · Accuracy ↗

1 more reported result

71.39 % · Gemini 2.5 Flash Lite (9/25) (Thinking) · LiveCodeBench · Accuracy · Vals AI · LiveCodeBench ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

There are not enough shared scored settings to identify close comparisons reliably.

Filter comparisons to this model →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. AIME — evaluation methodology and results ↗

    Vals AI · 2026

  2. CorpFin — evaluation methodology and results ↗

    Vals AI · 2026

  3. GPQA Diamond — evaluation methodology and results ↗

    Vals AI · 2026

  4. LiveCodeBench — evaluation methodology and results ↗

    Vals AI · 2026

  5. LegalBench — evaluation methodology and results ↗

    Vals AI · 2026

  6. MedCode — evaluation methodology and results ↗

    Vals AI · 2026

  7. MedQA — evaluation methodology and results ↗

    Vals AI · 2026

  8. MedScribe — evaluation methodology and results ↗

    Vals AI · 2026

  9. MGSM — evaluation methodology and results ↗

    Vals AI · 2026

  10. MMLU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  11. MMMU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  12. MortgageTax — evaluation methodology and results ↗

    Vals AI · 2026

  13. SAGE — evaluation methodology and results ↗

    Vals AI · 2026

  14. TaxEval — evaluation methodology and results ↗

    Vals AI · 2026

Names used in source tables

Gemini 2.5 Flash Lite (9/25) (Nonthinking) · Gemini 2.5 Flash Lite (9/25) (Thinking)