List your company

Question Answering on MMLU (Accuracy)

92AccuracyGPT-4.1
17.6436.94556.2575.555Oct 5, 2025Oct 29, 2025Nov 23, 2025Dec 17, 2025Jan 11, 2026Feb 4, 2026Mar 1, 2026

Evaluation results

MethodLinks
GPT-4.12025.1092
Claude Sonnet 42025.1089.5
DOCTOR-R12025.1085
Baichuan-M2-32B2025.1084
Evo 8BShots=5, E2E Latency (s)=8.6, Inference Speed (tokens/s)=522026.0276.8
Qwen2.5 7BShots=5, E2E Latency (s)=8.1, Inference Speed (tokens/s)=462026.0274.5
OriginBase Model=Qwen-3-8B2026.0273.01
FULLModel=Qwen3-8b, Zero-shot protocol=true2025.1272.97
DoctorAgent-RL2025.1072.5
Qwen3-8Bbase_model=true2025.1070
BD3-LM 7BShots=5, E2E Latency (s)=14.2, Inference Speed (tokens/s)=282026.0270
KUDABase Model=Qwen-3-8B2026.0268.92
SPONModel=Qwen3-8b, Zero-shot protocol=true2025.1268.79
TEALModel=Qwen3-8b, Zero-shot protocol=true2025.1268.78
FULLModel=Llama3-8b, Zero-shot protocol=true2025.1268.05
UltraMedical-8B2025.1067.5
LLaDA 8BShots=5, E2E Latency (s)=21.8, Inference Speed (tokens/s)=162026.0265.6
LLaMA3 8BShots=5, E2E Latency (s)=7.4, Inference Speed (tokens/s)=582026.0265.1
MDLM 7BShots=5, E2E Latency (s)=18.9, Inference Speed (tokens/s)=222026.0264
OriginBase Model=Llama-3.1-8B2026.0263.31
SPONModel=Llama3-8b, Zero-shot protocol=true2025.1260.95
KUDABase Model=Llama-3.1-8B2026.0260.64
Med42-v2-8B2025.1060.5
RMUBase Model=Qwen-3-8B2026.0260.02
TEALModel=Llama3-8b, Zero-shot protocol=true2025.1259.69
FULLModel=Mistral-v2-7B, Zero-shot protocol=true2025.1259.05
BaseModel=Mistral-7B, Zero-shot=true2026.0359
BaseModel=Zephyr-7B-β, Zero-shot=true2026.0358.5
RMUBase Model=Llama-3.1-8B2026.0258.33
OriginBase Model=Zephyr-7B-Beta2026.0258.13
RMUModel=Mistral-7B, Zero-shot=true2026.0357.7
ASUModel=Zephyr-7B-β, Zero-shot=true2026.0357.5
ASUModel=Mistral-7B, Zero-shot=true2026.0357.2
KUDABase Model=Zephyr-7B-Beta2026.0257.14
RMUModel=Zephyr-7B-β, Zero-shot=true2026.0357
RMUBase Model=Zephyr-7B-Beta2026.0256.9
SPONModel=Mistral-v2-7B, Zero-shot protocol=true2025.1255.78
TEALModel=Mistral-v2-7B, Zero-shot protocol=true2025.1255.37
SCRUBModel=Zephyr-7B-β, Zero-shot=true2026.0351.2
SimNPOBase Model=Zephyr-7B-Beta2026.0248.26
ARD 7BShots=5, E2E Latency (s)=32.5, Inference Speed (tokens/s)=122026.0246.2
NPOBase Model=Zephyr-7B-Beta2026.0245.71
LLMUModel=Zephyr-7B-β, Zero-shot=true2026.0344.7
SSDModel=Zephyr-7B-β, Zero-shot=true2026.0340.7
GABase Model=Zephyr-7B-Beta2026.0233.02
Gemini-2.5-Flash2025.1020.5