LiveBench 2.0 Design

Design and layout tasks across presentation, PDF and document outputs; visual and structural checks. LiveBench 2.0, refreshed with new questions each release to limit contamination.

CategoryFrontend
MetricScore (%)
SourceLiveBench (original)
LicenseNone stated
Benchmark pagelivebench.ai
StatusScored
Weight0.9 within Frontend
FitDiscrimination 0.62, noise 0.10

Results

#ModelProviderResultDate
1 Claude Fable 5.2 Anthropic 85.1% 2026-11-06
2 Claude Opus 5.5 Anthropic 83.3% 2026-11-06
3 Claude Fable 5.1 Anthropic 81.6% 2026-11-06
4 Claude Sonnet 5.5 Anthropic 81.3% 2026-11-06
5 Gemini 4 Carbon Google 80.2% 2026-11-06
6 GPT-6 Astra OpenAI 80.0% 2026-11-06
7 DeepSeek V4.1 Flash 1017 DeepSeek 79.8% 2026-11-06
8 GPT-6.1 Terra OpenAI 79.7% 2026-11-06
9 GPT-6.1 Sol OpenAI 79.6% 2026-11-06
10 Claude Haiku 5.5 Anthropic 79.3% 2026-11-06
11 Qwen3.8 Max Alibaba 76.6% 2026-11-06
12 Grok 4.6 xAI 76.3% 2026-11-06
13 Mistral Large 4 Preview Mistral 76.3% 2026-11-06
14 GPT-6 Luna OpenAI 76.1% 2026-11-06
15 Grok 4.7 xAI 75.0% 2026-11-06
16 Muse Spark 1.3 Meta 74.8% 2026-11-06
17 MiMo-V2.6-Pro Xiaomi 74.3% 2026-11-06
18 Kimi K3.1 Pro Moonshot AI 73.7% 2026-11-06
19 GLM-5.3-Flash Zhipu AI 73.7% 2026-11-06
20 Kimi K3 Moonshot AI 72.5% 2026-11-06
21 Qwen3.8 27B Alibaba 72.4% 2026-11-06
22 MiniMax-M3.1-Flash MiniMax 71.2% 2026-11-06
23 Muse Wonder Meta 70.9% 2026-11-06
24 GPT-5.6 Terra OpenAI 70.9% 2026-11-06
25 Step 5 Preview StepFun 70.8% 2026-11-06
26 Gemini 3.8 Flash Google 69.7% 2026-11-06
27 Gemini 4 Argon Google 69.2% 2026-11-06
28 Mistral Large 4 Mistral 60.7% 2026-11-06
29 Nemotron 3.5 Super VL 120B A12B NVIDIA 60.0% 2026-11-06
30 Muse Spark Open Meta 58.1% 2026-11-06
31 Gemma 4 31B IT Google 56.5% 2026-11-06
32 Inkling Thinking Machines 55.2% 2026-11-06