Benchmark Almanac

DAYJOB: Healthcare

Claude Opus 5.5 48.3% leads since September 2026
Not in model cards

Agents handle short requests from healthcare staff using charts, claims exports, policies and emails. An agent judge applies an expert rubric, and a pass needs 95% of criteria.

50 tasks across clinical decision support, pharmacy, payer work and compliance. The best models scored under 25% at launch.

Made by
Surge AI
Released
September 2026
Kind
Medicine
Level
Expert
Scored
Graded by a model
Answers
Agent task
Tasks
All public
Strongest model tested
GPT-6 Astra #3 on the capability index
Last model added
October 2026
Scores refreshed
November 12th, 2026
38
Entries on its own leaderboard, 37 models we track

Claude Opus 5.5 leads with 48.3%

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

38 entries. Scroll sideways for the rest.

  1. 48.3 Claude Opus 5.5
  2. 34.0 Claude Sonnet 5.5
  3. 31.2 Muse Wonder 1.0
  4. 31.2 Claude Fable 5.1
  5. 28.4 GPT-6 Astra
  6. 27.6 Grok 4.7
  7. 25.2 GPT-6.1 Sol
  8. 23.6 Gemini 4 Argon
  9. 22.0 Claude Opus 5
  10. 22.0 Muse Spark 1.3
  11. 14.8 Grok 4.6
  12. 11.2 Claude Fable 5
  13. 8.80 Muse Spark 1.2
  14. 7.20 GPT-6 Sol
  15. 6.40 Tencent Hy4 preview
  16. 6.00 DeepSeek V4.1 Flash 1017
  17. 4.80 Qwen 3.8 Max
  18. 4.40 GPT-5.6 Sol
  19. 4.40 GLM-5.3
  20. 4.00 Kimi K3
  21. 3.60 GPT-5.6 Terra
  22. 3.60 Mistral Large 4
  23. 2.80 Claude Sonnet 5
  24. 2.40 Gemini 3.7 Flash
  25. 2.00 Gemini 3.8 Flash
  26. 1.60 GPT-5.6 Luna
  27. 1.20 DeepSeek-V4-Pro
  28. 1.20 Tencent Hy3
  29. 1.20 GLM-5.3-Flash
  30. 1.20 GPT-6 Luna
  31. 1.20 Beam 1.0
  32. 0.40 Gemini 3.1 Pro
  33. 0.40 Nemotron 3 Ultra
  34. 0.40 Inkling
  35. 0.0 Mistral Large 3
  36. 0.0 DeepSeek-V4-Flash
  37. 0.0 Kimi K2.7 Code
  38. 0.0 Muse Glimmer
  1. 1 Claude Opus 5.5 adaptive/max 48.3%
  2. 2 Claude Sonnet 5.5 adaptive/max 34%
  3. 3 Muse Wonder 1.0 max 31.2%
  4. 4 Claude Fable 5.1 adaptive/max 31.2%
  5. 5 GPT-6 Astra max 28.4%
  6. 6 Grok 4.7 xhigh 27.6%
  7. 7 GPT-6.1 Sol max 25.2%
  8. 8 Gemini 4 Argon high 23.6%
  9. 9 Claude Opus 5 adaptive/max 22%
  10. 10 Muse Spark 1.3 max 22%
  11. 11 Grok 4.6 xhigh 14.8%
  12. 12 Claude Fable 5 adaptive/max 11.2%
  13. 13 Muse Spark 1.2 xhigh 8.80%
  14. 14 GPT-6 Sol max 7.20%
  15. 15 Tencent Hy4 preview high 6.40%
  16. 16 DeepSeek V4.1 Flash 1017 max 6.00%
  17. 17 Qwen 3.8 Max xhigh 4.80%
  18. 18 GPT-5.6 Sol max 4.40%
  19. 19 GLM-5.3 max 4.40%
  20. 20 Kimi K3 max 4.00%
  21. 21 GPT-5.6 Terra xhigh 3.60%
  22. 22 Mistral Large 4 high 3.60%
  23. 23 Claude Sonnet 5 adaptive/max 2.80%
  24. 24 Gemini 3.7 Flash high 2.40%
  25. 25 Gemini 3.8 Flash high 2.00%
  26. 26 GPT-5.6 Luna max 1.60%
  27. 27 DeepSeek-V4-Pro max 1.20%
  28. 28 Tencent Hy3 high 1.20%
  29. 29 GLM-5.3-Flash max 1.20%
  30. 30 GPT-6 Luna max 1.20%
  31. 31 Beam 1.0 max 1.20%
  32. 32 Gemini 3.1 Pro high 0.40%
  33. 33 Nemotron 3 Ultra 0.40%
  34. 34 Inkling max 0.40%
  35. 35 Mistral Large 3 0%
  36. 36 DeepSeek-V4-Flash max 0%
  37. 37 Kimi K2.7 Code max 0%
  38. 38 Muse Glimmer xhigh 0%

Source: Official DAYJOB: Healthcare leaderboard

#EntryScoreSettingDateSource
1 Claude Opus 5.5 48.3% adaptive/max Sep 2026 Official leaderboard
2 Claude Sonnet 5.5 34% adaptive/max Sep 2026 Official leaderboard
3 Muse Wonder 1.0 31.2% max Official leaderboard
4 Claude Fable 5.1 31.2% adaptive/max Sep 2026 Official leaderboard
5 GPT-6 Astra 28.4% max Sep 2026 Official leaderboard
6 Grok 4.7 27.6% xhigh Sep 2026 Official leaderboard
7 GPT-6.1 Sol 25.2% max Sep 2026 Official leaderboard
8 Gemini 4 Argon 23.6% high Sep 2026 Official leaderboard
9 Claude Opus 5 22% adaptive/max Jul 2026 Official leaderboard
10 Muse Spark 1.3 22% max Sep 2026 Official leaderboard
11 Grok 4.6 14.8% xhigh Aug 2026 Official leaderboard
12 Claude Fable 5 11.2% adaptive/max Jun 2026 Official leaderboard
13 Muse Spark 1.2 8.80% xhigh Aug 2026 Official leaderboard
14 GPT-6 Sol 7.20% max Sep 2026 Official leaderboard
15 Tencent Hy4 preview 6.40% high Aug 2026 Official leaderboard
16 DeepSeek V4.1 Flash 1017 6.00% max Oct 2026 Official leaderboard
17 Qwen 3.8 Max 4.80% xhigh Aug 2026 Official leaderboard
18 GPT-5.6 Sol 4.40% max Jul 2026 Official leaderboard
19 GLM-5.3 4.40% max Aug 2026 Official leaderboard
20 Kimi K3 4.00% max Jul 2026 Official leaderboard
21 GPT-5.6 Terra 3.60% xhigh Jul 2026 Official leaderboard
22 Mistral Large 4 3.60% high Oct 2026 Official leaderboard
23 Claude Sonnet 5 2.80% adaptive/max Jun 2026 Official leaderboard
24 Gemini 3.7 Flash 2.40% high Aug 2026 Official leaderboard
25 Gemini 3.8 Flash 2.00% high Sep 2026 Official leaderboard
26 GPT-5.6 Luna 1.60% max Jul 2026 Official leaderboard
27 DeepSeek-V4-Pro 1.20% max Apr 2026 Official leaderboard
28 Tencent Hy3 1.20% high Jul 2026 Official leaderboard
29 GLM-5.3-Flash 1.20% max Aug 2026 Official leaderboard
30 GPT-6 Luna 1.20% max Sep 2026 Official leaderboard
31 Beam 1.0 1.20% max Oct 2026 Official leaderboard
32 Gemini 3.1 Pro 0.40% high Feb 2026 Official leaderboard
33 Nemotron 3 Ultra 0.40% Jun 2026 Official leaderboard
34 Inkling 0.40% max Jul 2026 Official leaderboard
35 Mistral Large 3 0% Dec 2025 Official leaderboard
36 DeepSeek-V4-Flash 0% max Apr 2026 Official leaderboard
37 Kimi K2.7 Code 0% max Jun 2026 Official leaderboard
38 Muse Glimmer 0% xhigh Aug 2026 Official leaderboard

2 models have held number one on the official leaderboard since its release in September 2026

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic
Own leaderboard 2 leaders
Claude Opus 5.5 48.3%

Source: Official DAYJOB: Healthcare leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 5.5 Claude Fable 5.1 Claude Opus 5 Claude Fable 5 DeepSeek-V4-Pro Mistral Large 3 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Opus 5.5 Claude Fable 5.1

Source: Official DAYJOB: Healthcare leaderboard

No lab model card reports it yet

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.

More in Law, finance and medicine

All 179 in Law, finance and medicine
  • MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy. Qwen 3.8 Max 80.4%
  • HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
  • HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
  • FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy. Ling-3.1-flash 81.7%
  • KBL In use Models answer Korean legal knowledge and reasoning tasks and Korean bar exam questions. It has 7 knowledge tasks, 4 reasoning tasks and 2,510 bar exam items. Solar Pro 4 77.5%
  • KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. Solar Pro 4 93.2%