List your company
ResearchFormerly WizWand

Knowledge on GPQA (Score, TPF)

69.16ScoreLing-flash-2.0
19.146432.130745.11558.0993Nov 14, 2025Nov 28, 2025Dec 13, 2025Dec 27, 2025Jan 11, 2026Jan 25, 2026Feb 9, 2026
Updated 8mo ago

Evaluation results

MethodLinks
Ling-flash-2.02026.0269.16100
LLaDA2.1-flashInference Mode=Q Mode2026.0267.3237
LLaDA2.1-flashInference Mode=S Mode2026.0266.67395
LLaDA2.0-flash2026.0262.31329
Ling-mini-2.02026.0259.41—
Qwen3-30B-A3B-Inst-25072026.0254.14100
LLaDA2.1-minimode=Q Mode2026.0253.282.12
LLaDA2.1-minimode=S Mode2026.0248.363.62
Qwen3-8Bno think=true2026.0248.01—
LLaDA2.0-mini2026.0247.762.73
LFM2-1.2B# Total Params=1.2B, # Trained Tokens=11T2025.1131.47—
UM-190kshots=0-shot2025.1130.43—
UM-187kshots=0-shot2025.1129.85—
Llama-3.2-1B# Total Params=1.2B, # Trained Tokens=9T2025.1128.84—
UM-190kBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1128.77—
HelpSteershots=0-shot2025.1128.61—
LFM2-700M# Total Params=0.70B, # Trained Tokens=11T2025.1128.48—
UM-187kBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1128.35—
TuluDPOshots=0-shot2025.1128.27—
UltraFBshots=0-shot2025.1128.1—
ORPOshots=0-shot2025.1127.94—
UM-170kshots=0-shot2025.1127.92—
Qwen3-1.7B# Total Params=1.7B, # Trained Tokens=36T2025.1127.72—
SFTshots=0-shot2025.1127.6—
ORPOBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1127.52—
LFM2-350M# Total Params=0.35B, # Trained Tokens=11T2025.1127.46—
TuluDPOBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1127.43—
UltraFBBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1127.43—
HelpSteerBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1127.35—
UM-170kBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1127.35—
CodePrefBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1126.85—
SFTBackbone=Apertus-8B-SFT, Evaluation Protocol=0-shot2025.1125.67—
CodePrefshots=0-shot2025.1125.59—
Qwen3-0.6B# Total Params=0.6B, # Trained Tokens=36T2025.1122.14—
Gemma-3-1B# Total Params=1B, # Trained Tokens=2T2025.1121.07—