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

Knowledge on MMLU

85.93Accuracy

DictaLM 3.0 24B-Think

Updated 2d ago

Evaluation Results

Method	Links
DictaLM 3.0 24B-Think 2026.02		85.93
Llama-3.3-70B-Instruct 2026.01		85.7
Phi-4 2026.01		84.9
Mistral Small 3.1 2026.02		83.11
ReasonAny 2026.01		82.09
Linear 2026.01		81.9
Task Arithmetic 2026.01		81.89
DARE 2026.01		81.85
Qwen2.5-32B-Instruct (Safety) 2026.01		81.78
LED 2026.01		80.83
FuseLLM 2026.01		80.62
DeepSeek-R1-Distill-Qwen-32B (Reasoning) 2026.01		79.65
TIES 2026.01		79.52
ReasonAny 2026.01		78.86
Task Arithmetic 2026.01		78.79
Qwen2.5-14B-Instruct (Safety) 2026.01		78.73
Linear 2026.01		77.63
LED 2026.01		77.56
DARE 2026.01		76.94
FuseLLM 2026.01		76.25
Gemma 3 27B 2026.02		74.6
DeepSeek-R1-Distill-Qwen-14B (Reasoning) 2026.01		73.28
TIES 2026.01		72.93
Qwen3-4B 2025.11		72.25
Llama-3.1-8B-Instruct 2026.01		69.8
Foundation-Sec-8B-Reasoning 2026.01		68.3
Llama-Primus-Merged 2026.01		68.2
ReasonAny 2026.01		67.15
Safety 2026.01		67.14
Granite-4.0-H 2025.11		66.79
Foundation-Sec-8B-Instruct 2026.01		66
Task Arithmetic 2026.01		64.93
LFM2-8B-A1B 2025.11		64.84
LFM2-2.6B 2025.11		64.42
LED 2026.01		63.45
Linear 2026.01		63.25
DARE 2026.01		61.76
Llama-3.2-3B 2025.11		60.35
SmolLM3-3B 2025.11		59.84
FuseLLM 2026.01		59.6
Qwen3-1.7B 2025.11		59.11
Gemma-3-4B 2025.11		58.35
DPO 2026.01		57.02
GANPO 2026.01		56.93
TIES 2026.01		56.74
Base 2026.01		56.73
LFM2-1.2B 2025.11		55.23
Reasoning 2026.01		53.25
LFM2-700M 2025.11		49.9
CAD 2026.01		47.9
Baseline 2026.01		47.7
ActDec 2026.01		47.6
Disagreement 2026.01		47.2
ME 2026.01		47.1
GAME-LoRA 2026.01		46.9
Llama-3.2-1B 2025.11		46.6
Qwen3-0.6B 2025.11		44.93
TAIA 2024.05		43.73
Vanilla 2024.05		43.62
LFM2-350M 2025.11		43.43
TAIA 2024.05		42.39
LoRACL 2024.05		42.18
Vanilla 2024.05		41.67
L2 2024.05		41.3
EWC 2024.05		41.07
Gemma-3-1B 2025.11		40.08
Self-Distill 2024.05		39.87
Vanilla 2024.05		25.51
EWC 2024.05		23.67
L2 2024.05		23.63
LoRACL 2024.05		23.53