LiveBench 2.0 Data Analysis

Short data-analysis questions on small tables (consecutive events, joins, reformatting) with one correct answer. LiveBench 2.0, refreshed with new questions each release to limit contamination.

CategoryReasoning
MetricScore (%)
SourceLiveBench (original)
LicenseNone stated
Benchmark pagelivebench.ai
StatusScored
Weight0.8 within Reasoning
FitDiscrimination 0.41, noise 0.35

Results

#ModelProviderResultDate
1 Gemini 4 Carbon Google 86.5% 2026-11-06
2 Gemini 4 Argon Google 85.2% 2026-11-06
3 GPT-6 Astra OpenAI 83.0% 2026-11-06
4 GPT-6.1 Sol OpenAI 82.7% 2026-11-06
5 GPT-6.1 Terra OpenAI 82.3% 2026-11-06
6 Muse Wonder Meta 81.6% 2026-11-06
7 Claude Fable 5.2 Anthropic 81.5% 2026-11-06
8 Claude Opus 5.5 Anthropic 80.3% 2026-11-06
9 Claude Fable 5.1 Anthropic 80.3% 2026-11-06
10 Muse Spark 1.3 Meta 79.6% 2026-11-06
11 DeepSeek V4.1 Flash 1017 DeepSeek 79.5% 2026-11-06
12 GPT-5.6 Terra OpenAI 79.3% 2026-11-06
13 Kimi K3.1 Pro Moonshot AI 79.1% 2026-11-06
14 Kimi K3 Moonshot AI 78.7% 2026-11-06
15 Qwen3.8 Max Alibaba 78.4% 2026-11-06
16 Grok 4.7 xAI 76.9% 2026-11-06
17 Qwen3.8 27B Alibaba 76.6% 2026-11-06
18 Mistral Large 4 Preview Mistral 76.6% 2026-11-06
19 Mistral Large 4 Mistral 76.5% 2026-11-06
20 GLM-5.3-Flash Zhipu AI 76.4% 2026-11-06
21 Grok 4.6 xAI 75.3% 2026-11-06
22 GPT-6 Luna OpenAI 73.4% 2026-11-06
23 Inkling Thinking Machines 72.8% 2026-11-06
24 MiMo-V2.6-Pro Xiaomi 72.3% 2026-11-06
25 MiniMax-M3.1-Flash MiniMax 71.4% 2026-11-06
26 Nemotron 3.5 Super VL 120B A12B NVIDIA 64.6% 2026-11-06
27 Step 5 Preview StepFun 63.5% 2026-11-06
28 Muse Spark Open Meta 60.2% 2026-11-06
29 Claude Sonnet 5.5 Anthropic 59.5% 2026-11-06
30 Gemini 3.8 Flash Google 54.0% 2026-11-06
31 Gemma 4 31B IT Google 53.3% 2026-11-06
32 Claude Haiku 5.5 Anthropic 53.1% 2026-11-06