List your company
ResearchFormerly WizWand

Scientific Question Answering on GPQA (accuracy)

54.44AccuracySpurious
22.470430.770239.0747.3698May 28, 2026Jun 3, 2026Jun 10, 2026Jun 17, 2026Jun 23, 2026Jun 30, 2026Jul 7, 2026
Updated 3mo ago

Evaluation results

MethodLinks
SpuriousModel=Qwen/Qwen3.5-27B2026.0554.44
Spurious UniversalModel=Qwen/Qwen3.5-27B2026.0550.8
PromptWizardModel=Qwen/Qwen3.5-27B2026.0549.59
Qwen3 8BGen. Mode=AR, TPF=1.002026.0749.24
Nemotron-Labs-Diffusion-8BGen. Mode=AR, TPF=1.002026.0744.44
Nemotron-Labs-Diffusion-8BGen. Mode=Quad. SS, TPF=6.382026.0744.3
Nemotron-Labs-Diffusion-8BGen. Mode=Diff., TPF=2.572026.0743.94
Ministral3-8B Instruct-2512Gen. Mode=AR, TPF=1.002026.0742.87
Step-BackModel=Qwen/Qwen3.5-27B2026.0541.11
Nemotron-Labs-Diffusion-8BGen. Mode=Linear SS, TPF=5.992026.0740.4
SDAR-8B ChatGen. Mode=Diff., TPF=1.002026.0740.2
Least-to-MostModel=meta-llama/Llama-3.2-1B-Instruct2026.0540
Plan-and-SolveModel=meta-llama/Llama-3.2-1B-Instruct2026.0537.78
Least-to-MostModel=Qwen/Qwen3.5-27B2026.0537.78
Qwen2.5 7BGen. Mode=AR, TPF=1.002026.0737.12
Step-BackModel=Qwen/Qwen3.5-0.8B2026.0536.67
AnalogicalModel=Qwen/Qwen3.5-27B2026.0536.67
Zero-Shot CoTModel=Qwen/Qwen3.5-0.8B2026.0535.56
Plan-and-SolveModel=Qwen/Qwen3.5-27B2026.0535.56
Self-AskModel=Qwen/Qwen3.5-27B2026.0535.56
Self-AskModel=Qwen/Qwen3.5-0.8B2026.0534.44
Re-ReadingModel=allenai/Olmo-3-7B-Instruct2026.0534.44
AnalogicalModel=Qwen/Qwen3.5-0.8B2026.0533.33
Re-ReadingModel=meta-llama/Llama-3.2-1B-Instruct2026.0533.33
Self-AskModel=allenai/Olmo-3-7B-Instruct2026.0533.33
PromptWizardModel=allenai/Olmo-3-7B-Instruct2026.0533.33
SpuriousModel=allenai/Olmo-3-7B-Instruct2026.0533.33
Re-ReadingModel=Qwen/Qwen3.5-27B2026.0533.33
Zero-Shot CoTModel=Qwen/Qwen3.5-27B2026.0533.33
LLaDA-8B InstructGen. Mode=Diff., TPF=1.002026.0733.3
Dream-7B InstructGen. Mode=Diff., TPF=1.002026.0733
SpuriousModel=Qwen/Qwen3.5-0.8B2026.0532.22
SpuriousModel=meta-llama/Llama-3.2-1B-Instruct2026.0531.85
Zero-Shot CoTModel=allenai/Olmo-3-7B-Instruct2026.0531.11
SDAR-8B ChatGen. Mode=Diff., TPF=1.752026.0730.8
Re-ReadingModel=Qwen/Qwen3.5-0.8B2026.0530
AnalogicalModel=meta-llama/Llama-3.2-1B-Instruct2026.0530
Spurious UniversalModel=Qwen/Qwen3.5-0.8B2026.0528.93
Step-BackModel=meta-llama/Llama-3.2-1B-Instruct2026.0528.89
Least-to-MostModel=Qwen/Qwen3.5-0.8B2026.0527.78
AnalogicalModel=allenai/Olmo-3-7B-Instruct2026.0527.78
PromptWizardModel=Qwen/Qwen3.5-0.8B2026.0527.43
Plan-and-SolveModel=Qwen/Qwen3.5-0.8B2026.0526.67
Least-to-MostModel=allenai/Olmo-3-7B-Instruct2026.0526.67
Spurious UniversalModel=meta-llama/Llama-3.2-1B-Instruct2026.0526.3
Zero-Shot CoTModel=meta-llama/Llama-3.2-1B-Instruct2026.0525.56
Step-BackModel=allenai/Olmo-3-7B-Instruct2026.0525.56
Plan-and-SolveModel=allenai/Olmo-3-7B-Instruct2026.0525.56
Self-AskModel=meta-llama/Llama-3.2-1B-Instruct2026.0524.44
Spurious UniversalModel=allenai/Olmo-3-7B-Instruct2026.0524.07
PromptWizardModel=meta-llama/Llama-3.2-1B-Instruct2026.0523.7