Benchmark Almanac

HoloBench

Gemini 2.0 Flash Feb 2025 59.6% leads since February 2025
Not in model cards

Models answer questions by combining facts across database rows written as sentences. A judge gives full, partial, or zero credit for each expected result.

A later Megagon Labs paper tested five language models on 90 HoloBench examples with 100,000 tokens of context.

Made by
Megagon Labs
Released
October 2024
Kind
Reasoning over long documents
Level
School
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Generated
Refresh
One fixed set
Run by
Anyone
Strongest model tested
o3-mini #124 on the capability index
Last model added
February 2025
Scores refreshed
November 20th, 2026
14
Models scored in its paper

Gemini 2.0 Flash leads with 59.6% among models released by February 2025

Best score per model 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. Every run is in the next chart.

The newest model with a score here came out in February 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

Source: The HoloBench paper

#ModelScoreSettingDateSource
1 Gemini 2.0 Flash Feb 2025 59.6% Constant information amount (2k) Feb 2025 Official leaderboard
2 Gemini 2.0 Flash Thinking Jan 2025 57.2% Constant information amount (2k) Jan 2025 Official leaderboard
3 Llama 3.1-405B 56.5% Constant information amount (2k) Jul 2024 Official leaderboard
4 GPT-4o Aug 2024 55.9% Constant information amount (2k) Aug 2024 Official leaderboard
5 o3-mini 54.5% Constant information amount (2k) Jan 2025 Official leaderboard
6 DeepSeek-V3 52.6% Constant information amount (2k) Dec 2024 Official leaderboard
7 o1-mini 51.2% Constant information amount (2k) Sep 2024 Official leaderboard
8 Claude 3.5 Sonnet 46.4% Constant information amount (2k) Jun 2024 Official leaderboard
9 Llama 3.1-70B 45.7% Constant information amount (2k) Jul 2024 Official leaderboard
10 GPT-4o mini 45.3% Constant information amount (2k) Jul 2024 Official leaderboard
11 Gemini 1.5 Pro May 2024 40.6% Constant information amount (2k) May 2024 Official leaderboard
12 DeepSeek-R1 37.5% Constant information amount (2k) Jan 2025 Official leaderboard
13 Gemini 1.5 Flash May 2024 31.4% Constant information amount (2k) May 2024 Official leaderboard
14 Llama 3.1-8B 23% Constant information amount (2k) Jul 2024 Official leaderboard

Rankings by part

We read 5 parts from the benchmark’s paper, each ranking models on one part of the test, such as 4k tokens, 8k tokens and 16k tokens. Gemini 2.0 Flash Thinking leads 3 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

4k tokens

14 models we track have a score here.

  1. 1 Gemini 2.0 Flash Thinking Jan 2025 ↑1 69.5%
  2. 2 o3-mini ↑3 68.6%
  3. 3 Gemini 2.0 Flash Feb 2025 ↓2 66.2%
  4. 4 Llama 3.1-405B ↓1 64%
  5. 5 o1-mini ↑2 63.5%
  6. 6 GPT-4o Aug 2024 ↓2 62.5%
  7. 7 Llama 3.1-70B ↑2 60.5%
  8. 8 DeepSeek-V3 ↓2 59.5%
  9. 9 GPT-4o mini ↑1 51.5%
  10. 10 DeepSeek-R1 ↑2 51.1%

Scores at each effort level

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Gemini 2.0 Flash
Constant in… Constant in…
59.6%
Gemini 2.0 Flash Thinking
Constant in… Constant in…
57.2%
Llama 3.1-405B
Constant in… Constant in…
56.5%
GPT-4o
Constant in… Constant in…
55.9%
o3-mini
Constant in… Constant in…
54.5%
DeepSeek-V3
Constant in… Constant in…
52.6%
o1-mini
Constant in… Constant in…
51.2%
Claude 3.5 Sonnet
Constant in…
46.4%
Llama 3.1-70B
Constant in…
45.7%
GPT-4o mini
Constant in… Constant in…
45.3%
Gemini 1.5 Pro
Constant in… Constant in…
40.6%
DeepSeek-R1
Constant in…
37.5%

Source: The HoloBench paper

ModelEffortHarnessScoreCost per taskOutput tokens
Gemini 2.0 Flash Feb 2025 Constant information amount (2k) 59.6%
Gemini 2.0 Flash Thinking Jan 2025 Constant information amount (2k) 57.2%
Gemini 2.0 Flash Feb 2025 Constant information density (50%) 56.8%
Llama 3.1-405B Constant information amount (2k) 56.5%
GPT-4o Aug 2024 Constant information amount (2k) 55.9%
o3-mini Constant information amount (2k) 54.5%
DeepSeek-V3 Constant information amount (2k) 52.6%
o1-mini Constant information amount (2k) 51.2%
DeepSeek-V3 Constant information density (50%) 50.1%
Gemini 2.0 Flash Thinking Jan 2025 Constant information density (50%) 50.1%
Llama 3.1-405B Constant information density (50%) 50%
GPT-4o Aug 2024 Constant information density (50%) 49.2%
Claude 3.5 Sonnet Constant information amount (2k) 46.4%
o3-mini Constant information density (50%) 46%
Llama 3.1-70B Constant information amount (2k) 45.7%
Llama 3.1-70B Constant information density (50%) 45.5%
GPT-4o mini Constant information amount (2k) 45.3%
Claude 3.5 Sonnet Constant information density (50%) 44.8%
o1-mini Constant information density (50%) 44.8%
Gemini 1.5 Pro May 2024 Constant information amount (2k) 40.6%
GPT-4o mini Constant information density (50%) 39.2%
DeepSeek-R1 Constant information amount (2k) 37.5%
DeepSeek-R1 Constant information density (50%) 35.9%
Gemini 1.5 Pro May 2024 Constant information density (50%) 34.8%
Gemini 1.5 Flash May 2024 Constant information amount (2k) 31.4%
Gemini 1.5 Flash May 2024 Constant information density (50%) 28.3%
Llama 3.1-8B Constant information amount (2k) 23%
Llama 3.1-8B Constant information density (50%) 20.5%

3 models have held number one in the benchmark’s paper since its release in October 2024

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

Meta Google DeepMind
Benchmark paper Not tested since Feb 2025

Source: The HoloBench 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% 2024 2025 Near the ceiling (85%) Gemini 2.0 Flash Llama 3.1-405B Claude 3.5 Sonnet Gemini 1.5 Pro 0% 20% 40% 60% 80% 100% 2024 2025 Near the ceiling (85%) Gemini 2.0 Flash Claude 3.5 Sonnet

Source: Official HoloBench 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 21 months ago, in February 2025, for Gemini 2.0 Flash. 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 14 models we track.
  • Warning Newest score The newest came 21 months ago, in February 2025, for Gemini 2.0 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is o3-mini (#124).
  • Pass Headroom The best is 59.6%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 28 rows for models we track 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