RLTasksby ulam.ai
Models / Google

Google / MODEL EVIDENCE

Gemini 2.5 Flash Lite

5 benchmarks · 6 observations · 5 sources

Scores at a glance

5 benchmarks · large dot: Gemini 2.5 Flash Lite; small dots: closest models where reported

MedCode ↗27.11 %
19.7263.57

Gemini 2.5 Flash Lite (Nonthinking) · MedCode · Accuracy ↗

MedScribe ↗72.83 %
4.2791.43

Gemini 2.5 Flash Lite (Nonthinking) · MedScribe · Accuracy ↗

Closest models by score

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

4.0 percentile-point gap

3 of 3 shared benchmark comparisons

How similarity is measured

Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.

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. SealQA: Raising the Bar for Reasoning in Search-Augmented Language Models — published result table ↗

    Pham, Thinh, Nguyen, Nguyen, Zunjare, Pratibha, Chen, Weiyuan, Tseng, Yu-Min, Vu, Tu · 2025

  2. TextQuests: How Good are LLMs at Text-Based Video Games? — published result table ↗

    Phan, Long, Mazeika, Mantas, Zou, Andy, Hendrycks, Dan · 2025

  3. MedCode — evaluation methodology and results ↗

    Vals AI · 2026

  4. MedScribe — evaluation methodology and results ↗

    Vals AI · 2026

  5. SAGE — evaluation methodology and results ↗

    Vals AI · 2026

Names used in source tables

Gemini 2.5 Flash Lite (Nonthinking) · Gemini 2.5 Flash-Lite