Gemini-2.5-Flash-Thinking
Parameters
—
Architecture
Transformer
Released
—
License
—
About
No detailed description available for this model.
Benchmark Scores
| Benchmark | Score | Date | Standing |
|---|---|---|---|
|
MMLU-Pro
knowledge
|
70.65%
|
31.07.2025 |
#26/48
better than 45%
|
|
MMLU-Redux
knowledge
|
85.29%
|
31.07.2025 |
#14/30
better than 55%
|
|
GPQA Diamond
stem_reasoning
|
78.26%
|
31.07.2025 |
#57/80
better than 28%
|
|
SuperGPQA
knowledge
|
71.17%
|
31.07.2025 |
#13/28
better than 56%
|
|
HMMT Feb 25
stem_reasoning
|
32.00%
|
31.07.2025 |
#15/17
better than 12%
|
|
LiveCodeBench v6
stem_reasoning
|
55.93%
|
31.07.2025 |
#38/49
better than 23%
|
|
OJBench
stem_reasoning
|
30.69%
|
31.07.2025 |
#14/22
better than 38%
|
|
IFEval
instruction_following
|
88.78%
|
31.07.2025 |
#14/26
better than 44%
|
|
MMLU-ProX
multilingual
|
60.00%
|
31.07.2025 |
#10/19
better than 50%
|
|
INCLUDE
multilingual
|
100.00%
|
31.07.2025 |
#1/11
better than 100%
|
|
PolyMATH
multilingual
|
20.45%
|
31.07.2025 |
#9/11
better than 20%
|
|
AIME 2025
stem_reasoning
|
60.23%
|
31.07.2025 |
#16/24
better than 35%
|
What is a good MMLU-Pro score? What is a good MMLU-Redux score? What is a good GPQA Diamond score?
No comments yet
Be the first to share your thoughts!