Benchmark Signal

The LLM Leaderboard

Every major AI model scored out of 10, ranked by aggregated public benchmarks — independent, transparent, and updated regularly.

49
Models ranked
1–10
Score scale
2026-06
Last updated
AI model leaderboard, ranked by overall score out of 10.
# Model Score Category Open
1 Claude Opus 4.8 Anthropic 9.7 Best Overall
2 Claude Opus 4.7 Anthropic 9.6 general
3 Claude Opus 4.5 Anthropic 9.5 general
4 GPT-5.5 Pro OpenAI 9.5 Reasoning
5 GPT-5.5 OpenAI 9.4 Best Overall
6 Claude Opus 4.8 Fast Anthropic 9.3 Fast
7 GPT-5 OpenAI 9.3 general
8 Gemini Ultra 2 Google DeepMind 9.2 Best Overall
9 GPT-5 Turbo OpenAI 9.1 general
10 DeepSeek V4 Pro DeepSeek 9.0 general
11 Qwen3.7 Max Alibaba Cloud 9.0 Best Open-Weight
12 Qwen3.6 Max Preview Alibaba Cloud 8.9 general
13 Claude Sonnet 4 Anthropic 8.8 general
14 Llama 4 405B Meta 8.7 Best Open-Weight
15 GPT-5.5 Instant OpenAI 8.6 Best Cheap Model
16 DeepSeek R1 DeepSeek 8.5 Best Reasoning
17 DeepSeek V4 Flash DeepSeek 8.5 general
18 Gemini Pro 2 Google DeepMind 8.5 Best Multimodal
19 Grok 4.3 xAI 8.5 general
20 MiniMax M3 MiniMax 8.5 general
21 Nemotron 3 Ultra NVIDIA 8.5 Intelligent
22 Qwen3.7 Plus Qwen 8.5 Image Input
23 Qwen3.6 35B A3B Alibaba Cloud 8.5 Best Local Model
24 MiMo V2.5 Pro Xiaomi 8.5 Agentic
25 Gemini 3.5 Flash Google DeepMind 8.4 Best Cheap Model
26 Kimi K2.6 Moonshot 8.4 Coding
27 Mistral Large 3 Mistral AI 8.3 general
28 MiMo V2.5 Xiaomi 8.3 Intelligent
29 GLM 5.1 Z Ai 8.3 Coding
30 Grok 3 xAI 8.2 general
31 Qwen3.6 27B Alibaba Cloud 8.1 Best Local Model
32 Grok Build 0.1 xAI 8.0 Coding
33 Mistral Medium 3.5 Mistral 8.0 Intelligent
34 Qwen 3 72B Alibaba Cloud 8.0 general
35 Claude Haiku 4 Anthropic 7.8 Best Fast Model
36 Codestral Mistral AI 7.8 Best Coding
37 Llama 4 Scout Meta 7.8 general
38 DeepSeek V3 DeepSeek 7.5 general
39 GPT-4o OpenAI 7.5 general
40 LFM2.5-8B-A1B Liquid AI 7.5 Best Local Model
41 Step 3.7 Flash StepFun 7.5 general
42 Hy3 (Preview) Tencent 7.5 Agentic
43 Gemini Flash 2 Google DeepMind 7.2 Best Cheap Model
44 Grok 3 Mini xAI 7.0 general
45 Command R+ Cohere 6.8 general
46 Phi-4 Microsoft 6.5 general
47 Mistral Medium Mistral AI 6.2 general
48 Qwen 2.5 72B Alibaba Cloud 6.0 general
49 Llama 3.3 70B Meta 5.8 general

Press / to search, 0–9 to switch category, Esc to reset.