DAYJOB: Healthcare · Open Analysis

#	Model	Effort	Score	Cost / task	Harness
1	Opus 5.5	max	24.7%	–	–
2	GPT-6 Astra	max	11.6%	–	–
3	Fable 5.1	max	9.6%	–	–
4	Gemini 4	high	9.6%	–	–
5	Opus 5	max	8.4%	–	–
6	Grok 4.7	xhigh	8.4%	–	–
7	Muse Spark 1.3	max	7.8%	–	–
8	Fable 5	max	5.6%	–	–
9	GPT-6 Sol	max	3.6%	–	–
10	Grok 4.6	xhigh	2.8%	–	–
11	Muse Spark 1.2	xhigh	2.0%	–	–
12	Qwen 3.8 Max	xhigh	2.0%	–	–
13	GPT-5.6 Sol	max	1.6%	–	–
14	GPT-5.6 Terra	xhigh	1.6%	–	–
15	Kimi K3	max	1.2%	–	–
16	Gemini 3.8 Flash	high	0.8%	–	–
17	DeepSeek V4 Pro	max	0.4%	–	–
18	GLM 5.3	max	0.4%	–	–
19	GLM 5.3 Flash	max	0.4%	–	–
20	GPT-5.6 Luna	max	0.4%	–	–
21	Hy3	high	0.4%	–	–
22	Sonnet 5	max	0.0%	–	–
23	DeepSeek V4 Flash	max	0.0%	–	–
24	Gemini 3.1 Pro	high	0.0%	–	–
25	Gemini 3.7 Flash	high	0.0%	–	–
26	GPT-6 Luna	max	0.0%	–	–
27	Inkling	max	0.0%	–	–
28	Kimi K2.7 Code	max	0.0%	–	–
29	Mistral Large 3	none	0.0%	–	–
30	Nemotron 3 Ultra 550B A55B	max	0.0%	–	–