List your company
ResearchFormerly WizWand

Coding on HumanEval (mean score)

0.9695HumanEval Mean ScoreMinistral-3-R
-0.007580.2460850.499750.753415Jan 21, 2026Jan 30, 2026Feb 8, 2026Feb 17, 2026Feb 26, 2026Mar 7, 2026Mar 16, 2026
Updated 7mo ago

Evaluation results

MethodLinks
Ministral-3-R# Param.=8B2026.020.9695
Falcon-H1R# Param.=7B2026.020.9634
MiniCPM-SALA# Param.=9B2026.020.9512
Phi-42026.010.945
Qwen3# Param.=8B2026.020.939
Nemotron-Nano-v2# Param.=9B2026.020.939
Llama-3.3-70B-InstructParameters=70B2026.010.933
MiniCPM-4.1# Param.=8B2026.020.9146
Llama-3.1-8B-InstructParameters=8B2026.010.823
Foundation-Sec-8B-InstructParameters=8B2026.010.823
Llama-Primus-Merged2026.010.817
Foundation-Sec-8B-ReasoningParameters=8B2026.010.799
Qwen2.5-7B-InstBase=Base, Verifier=-2026.010.793
DARLBase=Base, Verifier=✗2026.010.75
VeriFreeBase=Base, Verifier=✗2026.010.72
RLVRBase=Base, Verifier=Rule2026.010.689
SimpleRL-ZooBase=Base, Verifier=Rule2026.010.671
DARLBase=Inst, Verifier=✗2026.010.64
RLPRBase=Inst, Verifier=✗2026.010.634
Llama3.1-8B-InstBase=Inst, Verifier=-2026.010.616
General ReasonerBase=Base, Verifier=Model2026.010.616
Oat-ZeroBase=Math, Verifier=Rule2026.010.61
SimpleRL-ZooBase=Math, Verifier=Rule2026.010.61
RLPRBase=Base, Verifier=✗2026.010.567
TTRLBase=Base, Verifier=Rule2026.010.506
MobileLLM-FlashNumber of Parameters=1.4B, Instruction Fine-Tuning (IFT)=true, Evaluation Protocol=user-assistant conversations2026.030.4634
MobileLLM-FlashNumber of Parameters=650M, Instruction Fine-Tuning (IFT)=true, Evaluation Protocol=user-assistant conversations2026.030.4512
Gemma3Number of Parameters=1B, Instruction Fine-Tuning (IFT)=true, Evaluation Protocol=user-assistant conversations2026.030.415
RLVRBase=Inst, Verifier=Rule2026.010.384
Qwen2.5-7BBase=-, Verifier=-2026.010.378
Llama3.2Number of Parameters=1B, Instruction Fine-Tuning (IFT)=true, Evaluation Protocol=user-assistant conversations2026.030.378
PRIMEBase=Math, Verifier=Rule2026.010.03