Bedside Bench: leaderboard

Physician-authored clinical cases in ten suites from differential diagnosis to medication safety, graded by an LLM judge against the authors' criteria; the score is the macro-average of suite means over 3 Fireworks runs of 500 cases. Answers are free-text case work, not exam multiple choice.

Metric: Macro-average of 10 suite means, 3 runs of 500 cases (%). Source: fireworks.ai. Saturation forecast: Estimated already saturated. 15 models tracked. Updated 16 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1Claude Opus 5Max94.6#20
2Muse Spark OpenMax94.3#81
3GPT-6.1 SolMax93.9#10
4GPT-6 AstraxHigh93.4#6
5Gemini 4 ArgonHigh92.7#12
6Kimi K3.1 ProMax91.46#29
7Kimi K3Max89.7#56
8Ember-1Max89.1#84
9GPT-6 SolxHigh88.1#26
10GPT-6 LunaxHigh87.2#99
11GLM-5.3Max86.9#68
12Gemini 3.8 FlashHigh86.5#36
13Grok 4.6xHigh83.3#45
14DeepSeek V4.1 FlashMax82.9#62
15DeepSeek V4 Pro (0813)Max80.6#90

No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.

Interactive version: theaggregate.ai/benchmark?slug=bedside-bench · How It Works · Data refreshed daily, snapshot 2026-11-16.