List your company

Coding Accuracy on LiveCodeBench

70AccuracyRSA
2.08819.71937.3554.981Nov 3, 2025Dec 9, 2025Jan 14, 2026Feb 19, 2026Mar 27, 2026May 2, 2026Jun 7, 2026

Evaluation results

MethodLinks
RSAType=test-time scaling (TTS)2026.0570
EVOLIB2026.0570
Best-of-NType=test-time scaling (TTS)2026.0567.5
DCType=test-time learning (TTL)2026.0565
BaseType=base sampling2026.0558.8
MemQModel=Gemma-4-E4B-it2026.0551.43
ExpRAGType=test-time learning (TTL)2026.0551.3
RAGModel=Gemma-4-E4B-it2026.0550.48
Self-RAGModel=Gemma-4-E4B-it2026.0549.52
MemPModel=Gemma-4-E4B-it2026.0549.52
Mem0Model=Gemma-4-E4B-it2026.0547.62
MemRLModel=Gemma-4-E4B-it2026.0545.71
No Mem.Model=Gemma-4-E4B-it2026.0544.76
RecursiveMASrecursion round=32026.0442.9
TextGradrecursion round=32026.0439.8
Dense Rollout BaselineModel=Qwen3-1.7B Thinking, Generation length cutoff=24K, Training protocol=One epoch on 21K filtered TACO, Sparsity=None2026.0639.7
Sparsity (0.86)Model=Qwen3-1.7B Thinking, Generation length cutoff=24K, Training protocol=One epoch on 21K filtered TACO, Sparsity Setting=0.862026.0639.33
Recursive-TextMASrecursion round=32026.0438.7
Single Agenttraining protocol=Full-SFT2026.0438.6
Single Agenttraining protocol=LoRA2026.0437.4
VanillaModel=Qwen 2.5 R1 7B, Compression Ratio (CR)=1, Sampling temperature=0.6, top-p=95%2025.1136.7
kvtcModel=Qwen 2.5 R1 7B, Compression Ratio (CR)=9-11, Variant=8x, Sampling temperature=0.6, top-p=95%2025.1136.5
DMSModel=Qwen 2.5 R1 7B, Compression Ratio (CR)=-, Variant=8x, Sampling temperature=0.6, top-p=95%2025.1133.4
kvtcModel=Qwen 2.5 R1 7B, Compression Ratio (CR)=18-21, Variant=16x, Sampling temperature=0.6, top-p=95%2025.1131.6
Mixture-of-Agents (MoA)recursion round=32026.0427
LoopLMrecursion round=32026.0424.9
General-ReasonerGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.0618.4
DARE + TIESGroup=Qwen3-4B-Base, Category=Merged Models2026.0616.63
VanillaModel=Qwen 2.5 R1 1.5B, Compression Ratio (CR)=1, Sampling temperature=0.6, top-p=95%2025.1116.4
kvtcModel=Qwen 2.5 R1 1.5B, Compression Ratio (CR)=9, Variant=8x, Sampling temperature=0.6, top-p=95%2025.1116.1
DMSModel=Qwen 2.5 R1 1.5B, Compression Ratio (CR)=-, Variant=8x, Sampling temperature=0.6, top-p=95%2025.1116.1
TIESGroup=Qwen3-4B-Base, Category=Merged Models2026.0615.66
RAMGroup=Qwen3-4B-Base, Category=Merged Models2026.0614.48
RESMERGEGroup=Qwen3-4B-Base, Category=Merged Models2026.0613.5
kvtcModel=Qwen 2.5 R1 1.5B, Compression Ratio (CR)=18, Variant=16x, Sampling temperature=0.6, top-p=95%2025.1113.3
TSV-MergeGroup=Qwen3-4B-Base, Category=Merged Models2026.0611.94
SpiralGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.069.98
TAGroup=Qwen3-4B-Base, Category=Merged Models2026.069.59
GRPOGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.065.28
Qwen3-4BGroup=Qwen3-4B-Base, Category=Base and Expert Models2026.064.7