Benchmark Almanac

RadLE 2.0

Claude Fable 5 758.0 leads since Jul 2026 Not in model cards

Models diagnose single-image radiology cases with a confidence score from 0 to 4, or say they do not know. A confidence-weighted index rewards right answers and penalizes confident errors.

The primary metric RadLE-C runs from 0 to 2000. It is larger than 1.0 and adds open-source and medical vision-language models, so scores do not compare.

Made by CRASH Lab , Koita Centre for Digital Health , Ashoka University Released Jul 2026 Medicine Tasks kept private Last model added Jul 2026 Scores refreshed 2 Nov 2026
Level
Expert Specialists with years of training, often over hours of work
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Short answer A number, word or short string matched to a key.
Leaderboard crashlab.in
16
Entries on its own leaderboard, 12 models we track

Claude Fable 5 leads with 758.0 among models released by Jul 2026

Best score per entry from the benchmark's own leaderboard, read on 2 Nov 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.

The newest model with a score here came out in Jul 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

16 entries. Scroll sideways for the rest.

  1. 758.0 Claude Fable 5
  2. 735.0 Muse Spark 1.1
  3. 665.0 GPT-5.6 Sol Pro
  4. 660.0 Gemini 3.1 Pro
  5. 541.0 OctoMed 7B
  6. 524.0 Grok 4.5
  7. 476.0 Nemotron 3 Omni
  8. 472.0 Qwen3.7-Plus
  9. 409.0 GLM-5V-Turbo
  10. 350.0 MiniMax-M3
  11. 336.0 Gemma 4 31B IT
  12. 284.0 Lingshu 32B
  13. 277.0 MedGemma 1.5 4B
  14. 263.0 Llama 4 Maverick
  15. 241.0 InternVL3.5-8B
  16. 181.0 Mistral Large 3
  1. 1 Claude Fable 5 758.0
  2. 2 Muse Spark 1.1 735.0
  3. 3 GPT-5.6 Sol Pro 665.0
  4. 4 Gemini 3.1 Pro 660.0
  5. 5 OctoMed 7B 541.0
  6. 6 Grok 4.5 524.0
  7. 7 Nemotron 3 Omni 476.0
  8. 8 Qwen3.7-Plus 472.0
  9. 9 GLM-5V-Turbo 409.0
  10. 10 MiniMax-M3 350.0
  11. 11 Gemma 4 31B IT 336.0
  12. 12 Lingshu 32B 284.0
  13. 13 MedGemma 1.5 4B 277.0
  14. 14 Llama 4 Maverick 263.0
  15. 15 InternVL3.5-8B 241.0
  16. 16 Mistral Large 3 181.0
#EntryScoreSettingDateSource
1 Claude Fable 5 758.0 Jun 2026 Official leaderboard
2 Muse Spark 1.1 735.0 Jul 2026 Official leaderboard
3 GPT-5.6 Sol Pro 665.0 Jul 2026 Official leaderboard
4 Gemini 3.1 Pro 660.0 Feb 2026 Official leaderboard
5 OctoMed 7B 541.0 Official leaderboard
6 Grok 4.5 524.0 Jul 2026 Official leaderboard
7 Nemotron 3 Omni 476.0 Official leaderboard
8 Qwen3.7-Plus 472.0 Jun 2026 Official leaderboard
9 GLM-5V-Turbo 409.0 Apr 2026 Official leaderboard
10 MiniMax-M3 350.0 Jun 2026 Official leaderboard
11 Gemma 4 31B IT 336.0 Apr 2026 Official leaderboard
12 Lingshu 32B 284.0 Official leaderboard
13 MedGemma 1.5 4B 277.0 Official leaderboard
14 Llama 4 Maverick 263.0 Apr 2025 Official leaderboard
15 InternVL3.5-8B 241.0 Aug 2025 Official leaderboard
16 Mistral Large 3 181.0 Dec 2025 Official leaderboard

Number one over time

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

Anthropic
Own leaderboard Not tested since Jul 2026
Claude Fable 5 758.0

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
0.00 200.0 400.0 600.0 800.0 2025 2026 0.00 200.0 400.0 600.0 800.0 2025 2026

No lab model card reports it yet

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.

Versions

Replaces Radiology's Last Exam (RadLE)

More in Law, finance and medicine

All 140 in Law, finance and medicine
  • 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%
  • 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 In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
  • 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%
  • PRBench In use Model answers 1,100 expert-written law and finance questions. LLM judge grades against weighted expert rubrics of 10-30 criteria each. Muse Spark 1.3 61.6%