BenchLeader
Menu
Best LLMsChoose a modelLeaderboardsModelsBenchmarksProvidersReleases15

LiveBench 2.0 Data Analysis

LiveBench 2.0 data analysis suite: short questions on small tables with one correct answer.

As of 21 Oct 2026, Gemini 4 Argon leads LiveBench 2.0 Data Analysis on BenchLeader with 85.2%, ahead of GPT-6 Astra at 83.0%, across 48 model configurations with a published result.

Published by
LiveBench
Category
Knowledge
Index weight
Reference only
Models
48
Data as of
21 Oct 2026

LiveBench (livebench.ai), White et al. 2024.

Not comparable with the earlier test of the same name, whose final scores stay on LiveBench 1.0 Data Analysis.

What the test looks like

LiveBench 2.0's data analysis suite: short questions on small tables (consecutive events, table joins, table reformatting), each with a single correct answer.

How it is scored

Percent correct on the current release, the equal-weight mean of its subcategories.

What to keep in mind

It shares a name with LiveBench 1.0's Data Analysis category but is a new question set, so the two are kept apart; 2.0 scores start on 19 October 2026.

Top 15
  1. 1Gemini 4 Argon85.2%
  2. 2GPT-6 Astra83.0%
  3. 3GPT-6.1 Sol82.7%
  4. 4Muse Wonder81.6%
  5. 5GPT-5.581.6%
  6. 6GPT-6 Sol81.2%
  7. 7Claude Fable 580.5%
  8. 8Claude Fable 5.180.3%
  9. 9Claude Opus 5.580.3%
  10. 10GPT-5.6 Sol79.8%
  11. 11Muse Spark 1.379.6%
  12. 12Deepseek v4 Flash Vision79.5%
  13. 13DeepSeek V4.1 Flash 101779.5%
  14. 14GPT-5.6 Terra79.3%
  15. 15DeepSeek V4.1 Flash79.3%
48 of 48
#
1Gemini 4 Argon85.2%63.32026-10-14
2GPT-6 Astra83.0%67.92026-10-14
3GPT-6.1 Sol82.7%63.32026-10-14
4Muse Wonder81.6%63.62026-10-14
5GPT-5.581.6%60.22026-10-14
6GPT-6 Sol81.2%–2026-10-14
7Claude Fable 580.5%67.12026-10-14
8Claude Fable 5.180.3%66.22026-10-14
9Claude Opus 5.580.3%61.92026-10-14
10GPT-5.6 Sol79.8%55.92026-10-14
11Muse Spark 1.379.6%58.32026-10-14
12Deepseek v4 Flash Vision79.5%–2026-10-14
13NewDeepSeek V4.1 Flash 101779.5%–2026-10-14
14GPT-5.6 Terra79.3%52.62026-10-14
15DeepSeek V4.1 Flash79.3%–2026-10-14
16Kimi K378.7%59.12026-10-14
17Gemini 3.1 Pro78.5%60.72026-10-14
18Qwen3.8 Maxmax78.4%63.32026-10-14
19GPT-5.6 Luna78.0%47.92026-10-14
20Grok 4.776.9%–2026-10-14
21Qwen3.8 27B76.6%54.52026-10-14
22Mistral Large 476.5%56.72026-10-14
23Muse Spark 1.276.5%–2026-10-14
24GLM 5.3 Flash76.4%62.72026-10-14
25MiniMax M376.2%56.52026-10-14
26Claude Sonnet 4.676.1%57.12026-10-14
27Grok 4.675.3%57.82026-10-14
28Claude Opus 574.6%65.42026-10-14
29Qwen3.8 Flash74.2%–2026-10-14
30GPT-6 Luna73.4%–2026-10-14
31Grok 4.573.0%58.92026-10-14
32MiMo-V2.6-Pro72.3%64.12026-10-14
33Claude Sonnet 571.7%56.02026-10-14
34NewMiniMax-M3.1-Flash71.4%61.92026-10-14
35GLM 5.370.2%–2026-10-14
36Gemini 3.7 Flash68.0%53.22026-10-14
37NewBeam 1.066.4%54.62026-10-14
38Claude Opus 4.866.0%60.12026-10-14
39Nemotron Super VL 3.564.6%–2026-10-14
40Step 5 Preview63.5%60.62026-10-14
41NewMuse Spark Open60.2%–2026-10-14
42Claude Sonnet 5.559.5%–2026-10-14
43Claude Haiku 4.559.3%45.32026-10-14
44Nemotron Ultra 354.5%–2026-10-14
45Gemini 3.8 Flash54.0%55.62026-10-14
46Gemma 4 31B53.3%48.32026-10-14
47Claude Haiku 5.553.1%–2026-10-14
48Nemotron Lightning 3.543.1%–2026-10-14

More knowledge benchmarks · All LLM benchmarks

Cite as: BenchLeader, “LiveBench 2.0 Data Analysis leaderboard”, https://www.benchleader.com/benchmarks/livebench2_data, data as of 21 Oct 2026.

LiveBench 2.0 Data Analysis: questions

What does LiveBench 2.0 Data Analysis measure?
LiveBench 2.0's data analysis suite: short questions on small tables (consecutive events, table joins, table reformatting), each with a single correct answer. Scores are reported in percent of tasks solved; higher is better.
Which AI model leads LiveBench 2.0 Data Analysis?
Gemini 4 Argon leads LiveBench 2.0 Data Analysis with 85.2% as of 21 Oct 2026, ahead of GPT-6 Astra at 83.0%.
How many models have LiveBench 2.0 Data Analysis results?
48 model configurations have a LiveBench 2.0 Data Analysis result on BenchLeader, all taken from LiveBench.
Who runs LiveBench 2.0 Data Analysis and how often is it updated?
LiveBench 2.0 Data Analysis is published by LiveBench. BenchLeader re-reads the published results once a day and records the date each result was published.
Does LiveBench 2.0 Data Analysis count toward the BenchLeader Index?
No. LiveBench 2.0 Data Analysis is shown for reference but left out of the composite index.