List your company
ResearchFormerly WizWand

General Knowledge and Language Understanding on MMLU (Accuracy)

92.39AccuracySafeSieve
42.168455.206768.24581.2833Aug 15, 2025Sep 22, 2025Oct 30, 2025Dec 7, 2025Jan 14, 2026Feb 21, 2026Mar 31, 2026
Updated 7mo ago

Evaluation results

MethodLinks
SafeSieveParadigm=post-prune, Base model=deepseek-V3-671B2025.0892.39
G-DesignerParadigm=pre-design, Base model=deepseek-V3-671B2025.0891.13
AgentPruneParadigm=post-prune, Base model=deepseek-V3-671B2025.0890.99
GPTSwarmParadigm=pre-design, Base model=deepseek-V3-671B2025.0890.52
AgentDropoutParadigm=post-prune, Base model=deepseek-V3-671B2025.0890.17
CoTParadigm=single, Base model=deepseek-V3-671B2025.0889.31
VanillaParadigm=single, Base model=deepseek-V3-671B2025.0887.97
G-DesignerParadigm=pre-design, Base model=gpt-4o-mini (~8B)2025.0887.2
AgentPruneParadigm=post-prune, Base model=gpt-4o-mini (~8B)2025.0883.3
GPTSwarmParadigm=pre-design, Base model=gpt-4o-mini (~8B)2025.0882.8
SafeSieveParadigm=post-prune, Base model=gpt-4o-mini (~8B)2025.0882.32
AgentDropoutParadigm=post-prune, Base model=gpt-4o-mini (~8B)2025.0880.01
CoTParadigm=single, Base model=gpt-4o-mini (~8B)2025.0878.43
VanillaParadigm=single, Base model=gpt-4o-mini (~8B)2025.0877.81
RTBackbone=Llama3-8B-instruct, Relative Compute=1x2026.0363.9
Llama3-8B-instructBackbone=Llama3-8B-instruct, Relative Compute=0x2026.0363.8
RT-EAT-LATBackbone=Llama3-8B-instruct, Relative Compute=9x2026.0361.3
Llama2-7B-chatBackbone=Llama2-7B-chat, Relative Compute=0x2026.0346.4
RTBackbone=Llama2-7B-chat, Relative Compute=1x2026.0345.6
RT-EAT-LATBackbone=Llama2-7B-chat, Relative Compute=9x2026.0345.4
RT-EATBackbone=Llama2-7B-chat, Relative Compute=9x2026.0344.8
R2D2Backbone=Llama2-7B-chat, Relative Compute=6558x2026.0344.1