List your company
ResearchFormerly WizWand

Meta-Evaluation on Aggregated Benchmarks (AIME, ARC, GSM8K, HE, MMLU, IT, RU, BFCL)

2.4Overall Average RankGPT-5 nano
1.8885.3448.812.256May 8, 2026
Updated 5mo ago

Evaluation results

MethodLinks
GPT-5 nano2026.052.4
Qwen3-8BParameters=8B2026.052.6
Qwen3-4BParameters=4B2026.054.2
gpt-oss-20bParameters=20B2026.054.5
Ministral-3-8BParameters=8B2026.055
gemma-3-12b-itParameters=12B, Instruction Tuned=true2026.055.1
EngGPT2-16B-A3BParameters=16B2026.058
Llama-3.1-8BParameters=8B2026.058.2
gemma-3-4b-itParameters=4B, Instruction Tuned=true2026.059.1
Moonlight-16B-A3BParameters=16B2026.059.9
Llama-3.2-3BParameters=3B2026.0510.4
LLaMAntino-3-8BParameters=8B2026.0511
Velvet-14BParameters=14B2026.0512.4
FastwebMIIA-7BParameters=7B2026.0513.4
deepseek-moe-16bParameters=16B2026.0514.5
Minerva-7BParameters=7B2026.0515.2