Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Skill retrieval on MedCalc-Bench
Loading...
92.3
nDCG@1
Qwen3-235B
26.78
43.79
60.8
77.81
Apr 27, 2026
nDCG@1
nDCG@10
Updated 1mo ago
Evaluation Results
Method
Method
Links
nDCG@1
nDCG@10
Qwen3-235B
Retrieval Stage=Rerank...
2026.04
92.3
92.4
Qwen3-32B
Retrieval Stage=Rerank...
2026.04
91.2
91.9
Mistral3.1-24B
Retrieval Stage=Rerank...
2026.04
91.1
91.9
Llama-3.3-70B
Retrieval Stage=Rerank...
2026.04
89.5
91.3
Llama-3.1-8B
Retrieval Stage=Rerank...
2026.04
86
88.9
Qwen3-4B
Retrieval Stage=Rerank...
2026.04
85.9
88.9
BGE
Retrieval Stage=First-...
2026.04
41.4
54.5
TF-IDF
Retrieval Stage=First-...
2026.04
37.5
52
Contriever
Retrieval Stage=First-...
2026.04
34.9
49.4
BM25
Retrieval Stage=First-...
2026.04
29.3
44.7
Hybrid
Retrieval Stage=First-...
2026.04
29.3
47.4
Feedback
Search any
task
Search any
task