List your company

Coding on MBPP

94.16ScoreMinistral-3-R
-3.0186422.2104347.439572.66857Jul 15, 2024Nov 4, 2024Feb 25, 2025Jun 17, 2025Oct 8, 2025Jan 28, 2026May 21, 2026

Evaluation results

MethodLinks
Ministral-3-R# Param.=8B2026.0294.16
Nemotron-Nano-v2# Param.=9B2026.0293.39
MiniCPM-4.1# Param.=8B2026.0291.05
Falcon-H1R# Param.=7B2026.0291.05
MiniCPM-SALA# Param.=9B2026.0289.11
Qwen3# Param.=8B2026.0281.32
SkillWeave#Params=1.42B, Backbone=Llama-3.2-1B-Instruct2026.0549.7
Twin-merging#Params=1.48B, Backbone=Llama-3.2-1B-Instruct2026.0549.4
Llama3.2-1B-Instruct#Params=1.15B, Backbone=Llama-3.2-1B-Instruct2026.0549
PEFT#Params=1.42B, Backbone=Llama-3.2-1B-Instruct2026.0548.2
FuseChat3.0#Params=1.48B, Backbone=Llama-3.2-1B-Instruct2026.0545.5
self-rewarding#Params=1.15B, Backbone=Llama-3.2-1B-Instruct2026.0544.6
Hybrid-KSAsetting=Train-from-scratch2026.0436.4
KSAsetting=Train-from-scratch2026.0435.8
Hybrid-GDNsetting=Train-from-scratch2026.0434.8
Hybrid-SCAsetting=Train-from-scratch2026.0431.6
Fullsetting=Train-from-scratch2026.0430.6
Hybrid-SWAsetting=Train-from-scratch2026.0430
Llama-3-70B-InstructType=Instruction-tuned2024.070.823
Qwen2-72B-InstructType=Instruction-tuned2024.070.802
Qwen1.5-110B-ChatType=Instruction-tuned2024.070.764
Mixtral-8x22B-InstructType=Instruction-tuned2024.070.759
Qwen1.5-72B-ChatType=Instruction-tuned2024.070.719