List your company

Question Answering on PubMedQA PQA-L (test)

87.08AccuracyGyan-4.4
9.953629.97685070.0232Oct 19, 2022May 22, 2023Dec 24, 2023Jul 27, 2024Feb 28, 2025Oct 2, 2025May 6, 2026

Evaluation results

MethodLinks
Gyan-4.4Category=Correct2026.0587.08418
BioGPT2022.1078.2—
MediSwift-XLSize=1.21B, Sparsity=dense2024.0376.8—
ClinicalAgentsCategory=Multi-Agents2026.0376.6—
ReConcileCategory=Multi-Agents2026.0376.4—
MediSwift-XLSize=605M, Sparsity=50%2024.0376.3—
Intern-S1Category=Single LLM2026.0375.8—
BioGPT-largeSize=1.54B2024.0375.5—
MediSwift-XLSize=302M, Sparsity=75%2024.0375.4—
GPT-4Shot-count=0-shot2024.0375.2—
MediSwift-LargeSize=510M, Sparsity=dense2024.0375.1—
Few-shot + CoTCategory=Single Agent2026.0374.6—
BioMedLMSize=2.70B2024.0374.4—
MediSwift-MedSize=302M, Sparsity=dense2024.0374.2—
MediSwift-LargeSize=255M, Sparsity=50%2024.0374.2—
MDAgentsCategory=Multi-Agents2026.0374.2—
MedChain-AgentsCategory=Multi-Agents2026.0374.2—
MediSwift-MedSize=151M, Sparsity=50%2024.0373.8—
BioGPT-medSize=345M2024.0373.6—
DRAGONSize=360M2024.0373.4—
PMC-LlamaSize=7.00B2024.0373.4—
MediSwift-LargeSize=128M, Sparsity=75%2024.0373.4—
ReActCategory=Single Agent2026.0373.2—
MediSwift-MedSize=76M, Sparsity=75%2024.0372.4—
MedAgentsCategory=Multi-Agents2026.0372.4—
ColaCareCategory=Multi-Agents2026.0372.4—
BioLinkBERTModel Variant=large2022.1072.2—
BioLinkBERT-largeSize=340M2024.0372.2—
RAGCategory=Single Agent2026.0371.8—
GPT-3.5Shot-count=0-shot2024.0371.6—
AutoGenCategory=Multi-Agents2026.0371.4—
BioLinkBERTModel Variant=base2022.1070.2—
BioLinkBERT-baseSize=110M2024.0370.2—
Llama-4-Maverick-17BCategory=Single LLM2026.0368.8—
GPT-5.2Category=Single LLM2026.0368.2—
MedGemma-4BCategory=Single LLM2026.0366.8—
FineMedLM-o1-8BCategory=Single LLM2026.0364.4—
BioELECTRA2022.1064.2—
BioELECTRASize=110M2024.0364.2—
DeepSeek-R1Category=Single LLM2026.0362—
Qwen3-VL-235BCategory=Single LLM2026.0356.2—
PubMedBERT2022.1055.8—
PubMedBERTSize=110M2024.0355.8—
HuaTuoGPT-o1-7BCategory=Single LLM2026.0355.2—
Gyan-4.4Category=Incorrect2026.0512.9262