BenchLeader
Menu
Best LLMsChoose a modelLeaderboardsModelsBenchmarksProvidersReleases12

LiveBench 1.0 Data Analysis

Data analysis category of LiveBench 1.0, frozen at its final release (25 June 2026). LiveBench 2.0's Data Analysis suite is a different test.

As of 24 Oct 2026, GPT-6 Astra leads LiveBench 1.0 Data Analysis on BenchLeader with 83.0%, ahead of GPT-6.1 Sol at 82.7%, across 67 model configurations with a published result.

Published by
LiveBench
Category
Knowledge
Index weight
Reference only
Models
67
Data as of
24 Oct 2026
Newest result
25 Jun 2026

The publisher has posted no new results on this board since 25 Jun 2026. BenchLeader re-reads it every day, so newer models appear here as soon as they are published.

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

This board is frozen. A different, newer test now carries its name, and its scores are on LiveBench 2.0 Data Analysis.

What the test looks like

LiveBench 1.0's data analysis tasks: reading tables, reformatting data, joining columns and predicting types, using recent datasets.

How it is scored

Percent correct, from LiveBench 1.0's final release (25 June 2026). LiveBench 2.0's Data Analysis suite is a different question set, so these scores no longer update.

What to keep in mind

A practical but narrow skill; models that are weak at structured output can lose points for formatting rather than reasoning.

Top 15
  1. 1GPT-6 Astra (max)83.0%
  2. 2GPT-6.1 Sol (max)82.7%
  3. 3GPT-6.1 Sol (xhigh)82.2%
  4. 4Muse Wonder (max)81.6%
  5. 5GPT-5.5 (xhigh)81.6%
  6. 6GPT-6 Sol (max)81.2%
  7. 7Claude Fable 5 (max)80.5%
  8. 8Claude Opus 5.5 (max)80.3%
  9. 9Claude Fable 5.1 (max)80.3%
  10. 10GPT-5.6 Sol (max)79.8%
  11. 11Claude Opus 5.5 (xhigh)79.8%
  12. 12Muse Spark 1.3 (xhigh)79.6%
  13. 13Deepseek v4 Flash Vision79.5%
  14. 14DeepSeek V4 Flash79.3%
  15. 15GPT-5.4 (xhigh)79.3%
67 of 67
#
1GPT-6 Astramax83.0%70.42026-06-25
2GPT-6.1 Solmax82.7%69.92026-06-25
3GPT-6.1 Solxhigh82.2%67.72026-06-25
4Muse Wondermax81.6%67.52026-06-25
5GPT-5.5xhigh81.6%66.12026-06-25
6GPT-6 Solmax81.2%66.42026-06-25
7Claude Fable 5max80.5%69.32026-06-25
8Claude Opus 5.5max80.3%72.62026-06-25
9Claude Fable 5.1max80.3%70.52026-06-25
10GPT-5.6 Solmax79.8%68.02026-06-25
11Claude Opus 5.5xhigh79.8%70.02026-06-25
12Muse Spark 1.3xhigh79.6%64.92026-06-25
13Deepseek v4 Flash Vision79.5%–2026-06-25
14DeepSeek V4 Flash79.3%52.22026-06-25
15GPT-5.4xhigh79.3%64.72026-06-25
16GPT-5.6 Terramax79.3%64.42026-06-25
17DeepSeek V4.1 Flashmax79.3%61.22026-06-25
18DeepSeek V4 Pro79.2%53.52026-06-25
19Smaug Flash79.1%–2026-06-25
20Smaug Mini78.8%–2026-06-25
21Kimi K378.7%59.12026-06-25
22Claude Sonnet 5.5xhigh78.6%67.92026-06-25
23Gemini 3.1 Prohigh78.5%60.32026-06-25
24Qwen3.8 Maxmax78.4%63.32026-06-25
25Claude Opus 4.7xhigh78.3%57.42026-06-25
26GPT-5.2-Codex78.2%57.32026-06-25
27GPT-5.2high78.2%56.02026-06-25
28GPT-5.6 Lunamax78.0%59.92026-06-25
29Claude Sonnet 4.6medium78.0%–2026-06-25
30Grok 4.7xhigh76.9%61.62026-06-25
31Qwen3.8 27B76.6%54.62026-06-25
32Mistral Large 4high76.5%49.82026-06-25
33Muse Spark 1.2xhigh76.5%63.22026-06-25
34GLM 5.3 Flash76.4%62.62026-06-25
35MiniMax M376.2%56.42026-06-25
36Ox Alphamax75.8%–2026-06-25
37Claude Opus 5max74.5%68.82026-06-25
38Claude Opus 4.5high74.4%59.22026-06-25
39Qwen3.8-Flash-Next74.2%60.22026-06-25
40Grok 4.673.9%57.82026-06-25
41GLM 5.273.7%52.02026-06-25
42GPT-6 Lunamax73.4%60.02026-06-25
43Grok 4.573.0%59.02026-06-25
44Claude Haiku 5.5xhigh72.8%61.52026-06-25
45Muse Spark 1.1xhigh72.5%61.62026-06-25
46Qwen3.7max71.8%60.62026-06-25
47Claude Sonnet 5xhigh71.7%59.02026-06-25
48GPT-5.4 Minixhigh70.8%56.32026-06-25
49Grok Build 0.170.8%54.82026-06-25
50Qwen3.6 27B70.4%48.62026-06-25
51GLM 5.370.2%–2026-06-25
52Qwen3.6 Plus69.9%58.82026-06-25
53Claude Opus 4.6high69.9%62.72026-06-25
54Gemini 3.7 Flashhigh68.0%63.52026-06-25
55GPT-5.4 Nanoxhigh67.6%54.22026-06-25
56Claude Opus 4.8max66.0%63.32026-06-25
57Kimi K2.6thinking65.1%59.42026-06-25
58Gemini 3.5 Flashhigh64.9%63.12026-06-25
59Gemini 3.6 Flashhigh63.0%62.02026-06-25
60Kimi K2.7 Code62.7%55.82026-06-25

More knowledge benchmarks · All LLM benchmarks

Cite as: BenchLeader, “LiveBench 1.0 Data Analysis leaderboard”, data as of 24 Oct 2026, https://www.benchleader.com/benchmarks/livebench_data.

LiveBench 1.0 Data Analysis: questions

What does LiveBench 1.0 Data Analysis measure?
LiveBench 1.0's data analysis tasks: reading tables, reformatting data, joining columns and predicting types, using recent datasets. Scores are reported in percent of tasks solved; higher is better.
Which AI model leads LiveBench 1.0 Data Analysis?
GPT-6 Astra leads LiveBench 1.0 Data Analysis with 83.0% as of 24 Oct 2026, ahead of GPT-6.1 Sol at 82.7%.
How many models have LiveBench 1.0 Data Analysis results?
67 model configurations have a LiveBench 1.0 Data Analysis result on BenchLeader, all taken from LiveBench.
Who runs LiveBench 1.0 Data Analysis and how often is it updated?
LiveBench 1.0 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 1.0 Data Analysis count toward the BenchLeader Index?
No. LiveBench 1.0 Data Analysis is shown for reference but left out of the composite index.