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

General Knowledge on MMLU (Accuracy)

91.1MMLU General Knowledge Accuracy

DeepSeek V3.2

Updated 4d ago

Evaluation Results

Method	Links
DeepSeek V3.2 2025.12		91.1
GLM 4.6 2025.12		90.7
LongCat-Flash Chat 2025.12		89.7
LongCat-Flash Exp-Chat 2025.12		89.6
Kimi-K2 Base 2026.01		87.8
DeepSeek-V3.2 Exp Base 2026.01		87.8
Denser 2025.12		87.8
Process Supervision 2025.12		87.7
DeepSeek-V3.1 Base 2026.01		87.4
Reflection-CoT 2025.12		87.3
MiMo-V2-Flash Base 2026.01		86.7
Self-Verification 2025.12		86.4
Denser 2025.12		86.2
Tree-of-Thought 2025.12		86.1
Process Supervision 2025.12		85.9
Self-Consistency 2025.12		85.6
Reflection-CoT 2025.12		85.5
Qwen-3-30B-A3B 2026.02		85
Think-to-Think 2025.12		84.9
Self-Verification 2025.12		84.7
Llama-3.3-70B 2026.02		84.6
Tree-of-Thought 2025.12		84.3
Qwen-3-32B 2026.02		84
Self-Consistency 2025.12		83.8
Chain-of-Thought 2025.12		83.2
Think-to-Think 2025.12		83.1
Denser 2025.12		81.8
Process Supervision 2025.12		81.7
Chain-of-Thought 2025.12		81.5
Reflection-CoT 2025.12		81.3
Qwen-3-14B 2026.02		81.2
Gemma-3-27B 2026.02		80.4
Denser 2025.12		80.3
Denser 2025.12		80.2
Process Supervision 2025.12		80.2
OLMo-3.1-32B 2026.02		80.1
Tree-of-Thought 2025.12		80.1
Self-Verification 2025.12		80.1
Process Supervision 2025.12		80
Reflection-CoT 2025.12		79.8
Reflection-CoT 2025.12		79.6
Self-Consistency 2025.12		79.6
Denser 2025.12		79.1
Think-to-Think 2025.12		78.9
Self-Verification 2025.12		78.7
Process Supervision 2025.12		78.6
Self-Verification 2025.12		78.6
Tree-of-Thought 2025.12		78.6
Tree-of-Thought 2025.12		78.4
Reflection-CoT 2025.12		78.2
Self-Consistency 2025.12		78.1
Self-Consistency 2025.12		77.9
Self-Verification 2025.12		77.5
Chain-of-Thought 2025.12		77.4
Think-to-Think 2025.12		77.4
Mistral-3.2-24B 2026.02		77.3
Think-to-Think 2025.12		77.2
Tree-of-Thought 2025.12		77
Denser 2025.12		76.9
Qwen3-8B 2026.01		76.8
Molmo2-8B 2026.01		76.6
Self-Consistency 2025.12		76.5
Process Supervision 2025.12		76.2
Gemma-3-12B 2026.02		76.1
Chain-of-Thought 2025.12		75.9
Reflection-CoT 2025.12		75.8
Think-to-Think 2025.12		75.8
Chain-of-Thought 2025.12		75.8
Qwen3-4B 2025.12		75.78
Self-Verification 2025.12		75.2
Tree-of-Thought 2025.12		74.7
Self-Consistency 2025.12		74.3
Chain-of-Thought 2025.12		74.3
Qwen-2.5-7B 2026.01		74.2
Think-to-Think 2025.12		73.5
Qwen3-4B 2026.01		72.2
Molmo2-4B 2026.01		72.2
Chain-of-Thought 2025.12		72.1
PeCAN 2025.12		71.83
NSPO 2025.12		71.74
BFPO 2025.12		71.72
Qwen2.5-7B-Instruct 2025.12		71.71
MoCAN 2025.12		71.71
DPO-H 2025.12		71.66
DPO-Mix 2025.12		71.56
DPO-S 2025.12		71.35
SafeRLHF 2025.12		70.77
W-DOOR 2025.12		70.25
GRPO (RLVRR) 2026.01		70.2
EuroLLM-22B 2026.02		69.8
GRPO (RM) 2026.01		69.6
GRPO (GRM) 2026.01		69.5
Instruct 2026.01		69.4
Llama-3.1-8B 2026.01		69.4
OLMo-3-7B 2026.02		69.3
GRPO (BLEU) 2026.01		69.2
DPO 2026.01		69.1
OLMo3-7B-Instruct 2026.01		69.1
GRPO (RLPR) 2026.01		68.7
Llama-3.1-8B 2026.02		68.3

Showing 100 of 170 rows