Benchmark Almanac

ANAH

GPT-4 (snapshot unspecified) 87.1% tops its leaderboard Baichuan2-13B 42.2% leads the models we track
Not in model cards

Models detect unsupported claims in answers to English and Chinese questions. The benchmark provides annotations for individual claims.

Released
May 2024
Kind
Hallucination rates
Level
University
Scored
Checked automatically
Answers
Prediction
Tasks
All public
Measures
Capability
Language
English, Chinese
In and out
Text → Text
Interaction
Single turn
Items from
Generated
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Llama 2-70B #210 on the capability index
Last model added
September 2023
Scores refreshed
November 20th, 2026
14
Entries in its paper, 7 models we track

GPT-4 (snapshot unspecified) tops the leaderboard with 87.1% Baichuan2-13B leads with 42.2% among models released by September 2023

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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 Baichuan2-13B, at 42.2%. This view shows only models we track, each at its best.

The newest model with a score here came out in September 2023, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 GPT-4 (snapshot unspecified) 87.1%
  2. 2 ANAH-20B 80.5%
  3. 3 ANAH-7B 78.7%
  4. 4 InternLM2-20B 61.5%
  5. 5 Qwen-72B 58.3%
  6. 6 GPT-3.5 (snapshot unspecified) 48%
  7. 7 Baichuan2-13B 42.2%
  8. 8 Llama 2-70B 18.4%
  9. 9 Qwen-14B 14.9%
  10. 10 Llama 2-7B 13.8%
  11. 11 InternLM2-7B 12.4%
  12. 12 Baichuan 2-7B 9.63%
  13. 13 Llama 2-13B 8.55%
  14. 14 Qwen-7B 8.46%

Source: The ANAH paper

#EntryScoreSettingDateSource
1 GPT-4 (snapshot unspecified) 87.1% Official leaderboard
2 ANAH-20B 80.5% Official leaderboard
3 ANAH-7B 78.7% Official leaderboard
4 InternLM2-20B 61.5% Official leaderboard
5 Qwen-72B 58.3% Official leaderboard
6 GPT-3.5 (snapshot unspecified) 48% Official leaderboard
7 Baichuan2-13B 42.2% Sep 2023 Official leaderboard
8 Llama 2-70B 18.4% Jul 2023 Official leaderboard
9 Qwen-14B 14.9% Sep 2023 Official leaderboard
10 Llama 2-7B 13.8% Jul 2023 Official leaderboard
11 InternLM2-7B 12.4% Official leaderboard
12 Baichuan 2-7B 9.63% Sep 2023 Official leaderboard
13 Llama 2-13B 8.55% Jul 2023 Official leaderboard
14 Qwen-7B 8.46% Sep 2023 Official leaderboard
#ModelScoreSettingDateSource
1 Baichuan2-13B 42.2% Sep 2023 Official leaderboard
2 Llama 2-70B 18.4% Jul 2023 Official leaderboard
3 Qwen-14B 14.9% Sep 2023 Official leaderboard
4 Llama 2-7B 13.8% Jul 2023 Official leaderboard
5 Baichuan 2-7B 9.63% Sep 2023 Official leaderboard
6 Llama 2-13B 8.55% Jul 2023 Official leaderboard
7 Qwen-7B 8.46% Sep 2023 Official leaderboard

Number one over time

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

Baichuan
Benchmark paper Not tested since May 2024

Source: The ANAH paper

Scores over time

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

Plot with other benchmarks
In the benchmark’s paper Best so far
0% 20% 40% 60% 80% 100% Near the ceiling (85%) Baichuan2-13B Llama 2-70B 0% 20% 40% 60% 80% 100% Near the ceiling (85%) Baichuan2-13B Llama 2-70B

Source: Official ANAH 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 Factuality

All 164 in Factuality
  • SimpleVQA In use Models answer 2,025 short factual questions about images across 9 tasks and 9 topics. An LLM judge grades answers against fixed references. Qwen3.7-Plus 81.7%
  • HallusionBench In use Models answer 1,129 yes or no questions about 346 images, built in pairs to expose language hallucination and visual illusion. Scored by accuracy. Keye-VL-2.0-30B-A3B 73.5%
  • SimpleQA Verified In use 1,000 human-verified, de-duplicated short-answer factual questions measuring parametric knowledge. Gemini 3.1 Pro 77.5%
  • AA-Omniscience v1.1 In use Models answer 3,000 knowledge questions in six domains without tools and may abstain. The index adds correct answers and subtracts wrong ones. Inkling 2 Small 16.1
  • GaRAGe In use Models write long-form RAG answers to 2,366 questions with human-annotated grounding passages. Scoring checks whether answers use the relevant passages. Amazon Nova Pro 60.7%
  • Kimi Hallu Bench In use Models answer hard prompts drawn from practical work tasks. Fact-checking agents score the share of claims that are hallucinated. GPT-6 Astra 4.20%

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 3 years ago, in September 2023, for Qwen-7B. 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 (14 entries).
  • Pass Models scored 7 models we track, plus 7 other entries on its leaderboard.
  • Warning Newest score The newest came 3 years ago, in September 2023, for Qwen-7B.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Llama 2-70B (#210).
  • Pass Headroom The best is 42.2%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 7 rows for models we track and 7 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