Benchmark Almanac

Solar Open 100B

Solar Open 100B places in the top three on KMMLU.

Upstage Released December 31st, 2025 Family Solar Open The lab's model card Compare with others
28
Benchmarks with a score
0
Ranked first, out of five or more models
1
In the top three
Upstage Dec 2025

Solar Open 100B

Open weights Mid-size

135.3 #228 of 404
Self-Reported Index, from the lab's own numbers. No Epoch Capabilities Index score yet.
  • IFBench 53.7% #61 /80
  • AA-LCR (Artificial Analysis Long Context Reasoning) 35% #36 /38
  • KMMLU-Pro 64% #7 /7

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
MMLU 88.2% #28 of 131 Lab model card
Arena-Hard v2.0 74.8% #7 of 24 Lab model card
KMMLU 73% #2 of 6 Lab model card
IFEval 88% #54 of 135 Lab model card
MMLU-Pro 80.4% #79 of 167 Lab model card
Ko-Arena-Hard v2 (Upstage) 79.9% #1 of 2 Lab model card
IFBench 53.7% #61 of 80 Lab model card
WritingBench 75.1 #12 of 14 Lab model card
Humanity's Last Exam 10.5% #140 of 154 Lab model card
AA-LCR (Artificial Analysis Long Context Reasoning) 35% #36 of 38 Lab model card
KMMLU-Pro 64% #7 of 7 Lab model card
CLIcK 78.9% #4 of 4 Lab model card
HAE-RAE Bench 73.3% #3 of 3 Lab model card
KoBALT 44.3% #3 of 3 Lab model card
Ko-IFEval (Upstage) 87.5% #1 of 1 Lab model card

Math and science

BenchmarkScoreRankSource
AIME 2024 91.7% #12 of 88 Lab model card
AIME 2025 84.3% #76 of 143 Lab model card
GPQA Diamond 68.1% #191 of 302 Lab model card
HMMT 2025 73.3% #44 of 54 Lab model card
HMMT Nov 2025 80% #24 of 28 Lab model card
HRM8K 87.6% #3 of 3 Lab model card
Ko-AIME 2025 (Upstage) 80% #2 of 2 Lab model card
Ko-AIME 2024 (Upstage) 80.3% #1 of 1 Lab model card

Coding

BenchmarkScoreRankSource
LiveCodeBench v6 74.2% #47 of 90 Lab model card

Agents and tools

BenchmarkScoreRankSource
tau2-bench 55.8% #70 of 79 Lab model card

Professional work

BenchmarkScoreRankSource
KBL 65.5% #4 of 4 Lab model card
KorMedMCQA 84.4% #4 of 4 Lab model card
KBank-MMLU (Upstage) 65.5% #3 of 3 Lab model card

Its lab's numbers put it at 135.3, number 228 of 404

The Self-Reported Index fits one score to the 14 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 133.1 to 137.5. How we compute it.

Its launch post

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

Solar-Open-100B model card (Hugging Face) Dec 31, 2025

Its system card

Every result in the longer documents the lab published with it. 3 of them aren't in the launch post.

Solar Open Technical Report Jan 11, 2026

Capabilities