Benchmark Almanac

LongHealth

Mixtral-8x7B-Instruct-v0.1 55% tops its leaderboard GPT-3.5 Turbo Nov 2023 51.7% leads the models we track
Not in model cards

Models answer 400 questions about long clinical records from 20 fictional patients. Tasks test retrieval, sorting and recognition of missing information.

The January 2024 paper compares nine open models and GPT-3.5 Turbo. Mixtral-8x7B-Instruct led single-patient information retrieval at 73% accuracy.

Released
January 2024
Kind
Reasoning over long documents
Level
University
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
GPT-3.5 Turbo Nov 2023 #201 on the capability index
Last model added
December 2023
Scores refreshed
November 20th, 2026
10
Entries in its paper, 2 models we track

Mixtral-8x7B-Instruct-v0.1 tops the leaderboard with 55% GPT-3.5 Turbo leads with 51.7%

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper reports no overall accuracy, so this is the plain mean of the three task accuracies. 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 best model we track is GPT-3.5 Turbo, at 51.7%. This view shows only models we track, each at its best.

  1. 1 Mixtral-8x7B-Instruct-v0.1 16,000-token prompt 55%
  2. 2 GPT-3.5 Turbo Nov 2023, 16,000-token prompt 51.7%
  3. 3 Mistral-7B-Instruct-v0.2 16,000-token prompt 49.7%
  4. 4 Yi-34B-200k 16,000-token prompt 46%
  5. 5 vicuna-13b-v1.5-16k 16,000-token prompt 34%
  6. 6 Yi-6B-200K 16,000-token prompt 28.3%
  7. 7 zephyr-7b-beta-16k 16,000-token prompt 27%
  8. 8 longchat-13b-16k 16,000-token prompt 21.7%
  9. 9 vicuna-7b-v1.5-16k 16,000-token prompt 21.7%
  10. 10 longchat-7b-v1.5-32k 16,000-token prompt 20.7%

Source: The LongHealth paper

#EntryScoreSettingDateSource
1 Mixtral-8x7B-Instruct-v0.1 55% 16,000-token prompt Official leaderboard
2 GPT-3.5 Turbo Nov 2023 51.7% 16,000-token prompt Nov 2023 Official leaderboard
3 Mistral-7B-Instruct-v0.2 49.7% 16,000-token prompt Dec 2023 Official leaderboard
4 Yi-34B-200k 46% 16,000-token prompt Official leaderboard
5 vicuna-13b-v1.5-16k 34% 16,000-token prompt Official leaderboard
6 Yi-6B-200K 28.3% 16,000-token prompt Official leaderboard
7 zephyr-7b-beta-16k 27% 16,000-token prompt Official leaderboard
8 longchat-13b-16k 21.7% 16,000-token prompt Official leaderboard
9 vicuna-7b-v1.5-16k 21.7% 16,000-token prompt Official leaderboard
10 longchat-7b-v1.5-32k 20.7% 16,000-token prompt Official leaderboard
#ModelScoreSettingDateSource
1 GPT-3.5 Turbo Nov 2023 51.7% 16,000-token prompt Nov 2023 Official leaderboard
2 Mistral-7B-Instruct-v0.2 49.7% 16,000-token prompt Dec 2023 Official leaderboard

Scores over time

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

In the benchmark’s paper
0% 20% 40% 60% 80% 100% 2024 Near the ceiling (85%) GPT-3.5 Turbo 0% 20% 40% 60% 80% 100% 2024 Near the ceiling (85%) GPT-3.5 Turbo

Source: Official LongHealth leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Long context

All 115 in Long context
  • MRCR v2 In use Multi-round coreference: retrieve the n-th of several similar requests hidden in a long synthetic conversation, up to 1M tokens. Muse Spark 1.3 98.5%
  • LongBench v2 In use 503 multiple-choice questions over 8K-2M-word contexts (QA, dialogue, code repos, structured data). Accuracy. Inkling 2 Small 70.6%
  • GraphWalks In use A graph is given as an edge list in long context. The model answers BFS or parent queries, scored by F1. Claude Fable 5 99.96%
  • OpenAI MRCR v2 In use Models read a long synthetic chat that hides 2, 4 or 8 identical writing requests and must reproduce the response to the i-th one. Scored by mean string match ratio per context bin. Gemini 3.7 Flash 89%
  • CL-bench In use Model must learn new knowledge from a supplied long context (rules, procedures, data) and solve tasks. Strict LLM-verified rubrics. GPT-5.4 27.9%
  • AA-LCR 1.1 In use Models answer 100 questions that need reasoning across sets of real documents of about 100K tokens. An LLM equality checker with a fixed system prompt compares each answer with the key. Step 5 Preview 88.3%

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in December 2023, for Mistral-7B-Instruct-v0.2. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the benchmark’s paper (10 entries).
  • Pass Models scored 2 models we track, plus 8 other entries on its leaderboard.
  • Warning Newest score The newest came 2 years ago, in December 2023, for Mistral-7B-Instruct-v0.2.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-3.5 Turbo (#201).
  • Pass Headroom The best is 51.7%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 2 rows for models we track and 8 other entries at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check