Benchmark Almanac

Solar Pro 2

Solar Pro 2 ranks first on 2 of the 11 benchmarks with five or more models scored.

Upstage Released July 10th, 2025 Family Solar Pro The lab's model card Compare with others
15
Benchmarks with a score
2
Ranked first, out of five or more models
2
In the top three
Upstage Jul 2025

Solar Pro 2

Closed weights Flagship

131.1 #269 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
  • KMMLU 77% #1 /6
  • MMLU-Pro 76.5% #102 /167
  • AIME 2025 69.1% #114 /143

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Knowledge and language

BenchmarkScoreRankSource
KMMLU 77% #1 of 6 Lab model card
MMLU 88.2% #29 of 131 Lab model card
HAE-RAE Bench 86.2% #1 of 3 Lab model card
IFEval 87.8% #56 of 135 Lab model card
MMLU-Pro 76.5% #102 of 167 Lab model card
Arena-Hard v2.0 46.2% #22 of 24 Lab model card

Math and science

BenchmarkScoreRankSource
MATH-500 95.4% #38 of 70 Lab model card
AIME 2024 76.5% #50 of 88 Lab model card
AIME 2025 69.1% #114 of 143 Lab model card

Coding

BenchmarkScoreRankSource
HumanEval 98.8% #1 of 101 Lab model card
SWE-bench Verified 40.8% #144 of 155 Lab model card

Agents and tools

BenchmarkScoreRankSource
Berkeley Function Calling Leaderboard (BFCL) 63.9% #32 of 41 Lab model card

Professional work

BenchmarkScoreRankSource
KBL 75.5% #2 of 4 Lab model card
KBank-MMLU (Upstage) 71% #2 of 3 Lab model card
KorMedMCQA 87.4% #3 of 4 Lab model card

Its lab's numbers put it at 131.1, number 269 of 404

The Self-Reported Index fits one score to the 10 capability results Upstage reported for this model, on the scale of the Epoch Capabilities Index. Labs choose which benchmarks to show and how to run them, so read it as the lab's claim, not an independent measurement. The 90% interval runs from 128.7 to 133.4. How we compute it.

Its launch post

The benchmarks the lab chose to headline, with the scores it gave.