Clinical & Medical
167 cases · 29 rankedNumeric reasoning, extraction, contract-holding and safety, measured on the file you ship.
- 1=
Gemma 4 E4B ITiPhone 17 Pro+89.8% - 1=
Nanbeige4.2 3BiPhone 17 Pro+86.2% - 1=
LFM2.5 2.6B86.1% - 1=
Spark X2.6 4BiPhone 17 Pro+86.0% - 1=
Ternary Bonsai 2 4B82.8% - +1 tied
Wellness
160 cases · 29 rankedHolds a wellness app's output contract: format, voice, language, the user's numbers.
- 1=
Spark X2.6 4BiPhone 17 Pro+96.6% - 1=
Gemma 4 E4B ITiPhone 17 Pro+96.1% - 1=
Gemma 4 E2B ITiPhone 17 Pro+95.3% - 1=
Spark X2.5 4B94.3% - 1=
Granite 4.0 H Micro94.3% - +3 tied
Agentic Clinical
140 cases · 29 rankedCalls the right health tool, or none: asks instead of inventing, escalates red flags, carries doses exactly.
- 1=
MedPsy 4BiPhone 17 Pro+94.4% - 1=
LFM2.5 2.6B91.1% - 1=
Gemma 4 E4B ITiPhone 17 Pro+89.3% - 1=
NVIDIA Nemotron 3 Nano 4B88.9% - 1=
Spark X2.6 4BiPhone 17 Pro+87.6% - +1 tied
Multimodal Health
460 cases · 9 rankedReads synthetic health pictures and simulated phone photos, and knows when not to: reports, boxes, devices, charts, handwriting.
- 1=
Gemma 4 E4B ITiPhone 17 Pro+73.8% - 1=
LFM3 VL 2B70.4% - 1=
Ministral 3 3B Instruct 2512iPhone 17 Pro+69.8% - 1=
Qwen3.5 2B67.1% - 1=
LFM2.5 VL 3B65.5%
System One
993 cases · 55 ranked · 18 phone buildsModels making bounded clinical decisions: each answer is one of the options supplied with the case.
- 1=
Drex MaxLarger88.2% - 1=
mammoth-1Larger86.7% - 1=
d3-ultraLarger86.6% - 2
d1-ultraAPI only85.4% - 4=
d3-turboLarger84.3% - +1 tied