Our new X account is live! Follow @wizwand_team for updates

Mathematical Reasoning on MATH 500 (Accuracy (%))

100Accuracy (%)

Gemini-3-Pro

Updated 3d ago

Evaluation Results

Method	Links
Gemini-3-Pro 2026.02		100
GPT-5.2 2026.02		99
Claude-Sonnet 4.5 2026.02		99
DS-Qwen 14B 2025.05		92.8
CATS 2025.05		92.8
Qwen3-235B-A22B 2026.02		92.4
TwC (Ours) - Img & Txt 2026.02		92.3
E2E Caprese 2025.05		92
Layer-wise Caprese 2025.05		91
Layer-wise Caprese 2025.05		90.8
DeepSeek-R1 2026.02		90.4
DS-Qwen 7B 2025.05		90.2
TwC (Ours) - Only Image 2026.02		90
E2E Caprese 2025.05		90
GRIFFIN 2025.05		89.8
CATS 2025.05		89.2
E2E Caprese 2025.05		89.2
Layer-wise Caprese 2025.05		87
E2E Caprese 2025.05		85.4
Layer-wise Caprese 2025.05		84.8
s1.1-7B 2026.02		81.4
HALLUGUARD 2026.01		81
GRIFFIN 2025.05		80.6
DS-Qwen 1.5B 2025.05		79.4
Energy 2026.01		78
MIND 2026.01		77.1
Perplexity 2026.01		77.1
LN Entropy 2026.01		76.2
RACE 2026.01		75.1
Inside 2026.01		74.9
E2E Caprese 2025.05		74.8
SelfCheck GPT 2026.01		74
Layer-wise Caprese 2025.05		73.8
IO Prompt 2026.01		72.7
Semantic Ent. 2026.01		72.5
CATS 2025.05		72
Qwen3-8B 2026.02		71.94
FActScore 2026.01		71.6
Qwen3-8B 2026.02		70.23
TWI-1-Generated Photo 2026.02		70.2
Qwen3-4B 2026.02		69.73
Qwen3-4B 2026.02		69.55
P(true) 2026.01		67.1
Sora 2 2026.02		67
DGR SFT 2026.02		65.6
DGR SFT 2026.02		64.4
E2E Caprese 2025.05		60.4
Qwen3-4B + TTSV 2025.12		60.2
DGR SFT 2026.02		59
LLaMA3.1-8B 2026.02		56.21
Vanilla SFT 2026.02		54.3
LLaMA3.1-8B 2026.02		54.21
Qwen3-4B 2025.12		51.8
Layer-wise Caprese 2025.05		47.2
GRIFFIN 2025.05		42
Vanilla SFT 2026.02		20
Qwen2.5-7B-Instruct 2026.02		19.2
DREAMLLM 2026.02		12.6
Vanilla SFT 2026.02		10.2