Benchmark Almanac

Galileo LLM Hallucination Index

GPT-4 Jun 2023 0.79 leads since November 2023
Not in model cards
Version 2023

Models answer questions with and without retrieved context and generate long responses. Galileo judges factual accuracy and adherence to the supplied context.

Made by
Galileo
Released
November 2023
Kind
Hallucination rates
Level
University
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Measures
Propensity
Behavior
Hallucination
Language
English
In and out
Text → Text
Interaction
Single turn
Run by
Anyone
Strongest model tested
GPT-4 Jun 2023 #186 on the capability index
Last model added
November 2023
Scores refreshed
November 20th, 2026
11
Entries on its own leaderboard, 9 models we track

GPT-4 leads with 0.79 among models released by November 2023 GPT-4 leads with 0.79 among models released by November 2023

Best score per entry from the benchmark's own leaderboard, read on November 20th, 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. This view shows only models we track, each at its best.

The newest model with a score here came out in November 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 Jun 2023 0.79
  2. 2 GPT-3.5 Turbo Nov 2023 0.77
  3. 3 GPT-3.5 Turbo Jun 2023 0.75
  4. 4 Llama 2-70B 0.72
  5. 5 GPT-3.5 Turbo Instruct 0.71
  6. 6 Llama 2-13B 0.69
  7. 7 Zephyr 7B Beta 0.68
  8. 8 Mistral-7B-Instruct-v0.1 0.64
  9. 9 falcon-40b-instruct 0.63
  10. 10 Llama 2-7B 0.63
  11. 11 mpt-7b-instruct 0.50

Source: Official Galileo LLM Hallucination Index leaderboard

#EntryScoreSettingDateSource
1 GPT-4 Jun 2023 0.79 Jun 2023 Official leaderboard
2 GPT-3.5 Turbo Nov 2023 0.77 Nov 2023 Official leaderboard
3 GPT-3.5 Turbo Jun 2023 0.75 Jun 2023 Official leaderboard
4 Llama 2-70B 0.72 Jul 2023 Official leaderboard
5 GPT-3.5 Turbo Instruct 0.71 Sep 2023 Official leaderboard
6 Llama 2-13B 0.69 Jul 2023 Official leaderboard
7 Zephyr 7B Beta 0.68 Oct 2023 Official leaderboard
8 Mistral-7B-Instruct-v0.1 0.64 Sep 2023 Official leaderboard
9 falcon-40b-instruct 0.63 Official leaderboard
10 Llama 2-7B 0.63 Jul 2023 Official leaderboard
11 mpt-7b-instruct 0.50 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-4 Jun 2023 0.79 Jun 2023 Official leaderboard
2 GPT-3.5 Turbo Nov 2023 0.77 Nov 2023 Official leaderboard
3 GPT-3.5 Turbo Jun 2023 0.75 Jun 2023 Official leaderboard
4 Llama 2-70B 0.72 Jul 2023 Official leaderboard
5 GPT-3.5 Turbo Instruct 0.71 Sep 2023 Official leaderboard
6 Llama 2-13B 0.69 Jul 2023 Official leaderboard
7 Zephyr 7B Beta 0.68 Oct 2023 Official leaderboard
8 Mistral-7B-Instruct-v0.1 0.64 Sep 2023 Official leaderboard
9 Llama 2-7B 0.63 Jul 2023 Official leaderboard

Rankings by part

We read 3 parts from the benchmark’s own leaderboard: Q&A with RAG, Q&A without RAG and Long-Form Text Generation. Each ranks models on one part of the test. GPT-4 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.

Q&A with RAG

9 models we track have a score here.

  1. 1 GPT-4 Jun 2023 0.76
  2. 2 GPT-3.5 Turbo Jun 2023 ↑1 0.75
  3. 3 GPT-3.5 Turbo Nov 2023 ↓1 0.74
  4. 4 Zephyr 7B Beta ↑3 0.71
  5. 5 Llama 2-13B ↑1 0.68
  6. 6 Llama 2-70B ↓2 0.68
  7. 7 GPT-3.5 Turbo Instruct ↓2 0.68
  8. 8 Mistral-7B-Instruct-v0.1 0.67
  9. 9 Llama 2-7B 0.65

Number one over time

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

OpenAI
Own leaderboard Not tested since Nov 2023
GPT-4 0.79

Source: Official Galileo LLM Hallucination Index leaderboard

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
On the benchmark’s own leaderboard
0.65 0.70 0.75 0.80 2023 2024 0.65 0.70 0.75 0.80 2023 2024

Source: Official Galileo LLM Hallucination Index 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.

Versions

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

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (11 entries).
  • Pass Models scored 9 models we track, plus 2 other entries on its leaderboard.
  • Note Newest score The newest came 3 years ago, in November 2023, for GPT-3.5 Turbo. That’s expected for a test replaced by Galileo LLM Hallucination Index RAG Special.
  • Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-4 (#186).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 9 rows for models we track and 2 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