Skip to main content
Meo

LCR (long-context reasoning) — AI model leaderboard

AI models ranked by LCR (long-context reasoning), an aggregated third-party benchmark from artificial_analysis. Higher is better. Cross-referenced against our first-party meo scores and Effective Value (𝕍).

Ranking 78 models across the full field · as of 2026-06-07.

#ModelLabLCR (long-context reasoning)
1OpenAI: GPT-5.2-Codexopenai75.7%
2OpenAI: GPT-5openai75.6%
3OpenAI: GPT-5.1openai75.0%
4OpenAI: GPT-5.5openai74.3%
5MiniMax: MiniMax M3minimax74.0%
6OpenAI: GPT-5.3-Codexopenai74.0%
7OpenAI: GPT-5.4openai74.0%
8Xiaomi: MiMo-V2.5-Proxiaomi73.3%
9Google: Gemini 3.1 Pro Previewgoogle72.7%
10OpenAI: GPT-5.2 Chatopenai72.7%
11Anthropic: Claude Opus 4.7anthropic70.3%
12Qwen: Qwen3.6 Plusqwen69.7%
13Google: Gemini 3.5 Flashgoogle69.3%
14OpenAI: GPT-5.4 Miniopenai69.3%
15OpenAI: o3openai69.3%
16Qwen: Qwen3.7 Maxqwen69.0%
17OpenAI: GPT-5 Codexopenai69.0%
18MiniMax: MiniMax M2.7minimax68.7%
19Qwen: Qwen3.6 27Bqwen68.7%
20OpenAI: GPT-5 Miniopenai68.0%
21Anthropic: Claude Opus 4.8anthropic67.7%
22OpenAI: GPT-5.1-Codexopenai67.3%
23Qwen: Qwen3.5-122B-A10Bqwen66.7%
24DeepSeek: DeepSeek V4 Prodeepseek66.3%
25OpenAI: GPT-5.4 Nanoopenai66.0%
26Kwaipilot: KAT-Coder-Pro V2kwaipilot66.0%
27Google: Gemini 2.5 Progoogle66.0%
28Qwen: Qwen3.5 397B A17Bqwen65.7%
29Google: Gemini 3.1 Flash Litegoogle65.3%
30Qwen: Qwen3.7 Plusqwen65.0%
31xAI: Grok 4.3x-ai64.3%
32inclusionAI: Ring-2.6-1Tinclusionai64.3%
33StepFun: Step 3.7 Flashstepfun63.7%
34Qwen: Qwen3.6 35B A3Bqwen63.7%
35DeepSeek: DeepSeek V4 Flashdeepseek63.0%
36OpenAI: GPT-5.1-Codex-Miniopenai62.7%
37Z.ai: GLM 5.1z-ai62.3%
38Google: Gemma 4 31Bgoogle62.0%
39Mistral: Mistral Medium 3.5mistralai61.0%
40Z.ai: GLM 5V Turboz-ai61.0%
41OpenAI: GPT-4.1openai61.0%
42Z.ai: GLM 5 Turboz-ai60.7%
43OpenAI: o1openai59.3%
44Qwen: Qwen3.5-9Bqwen59.0%
45Anthropic: Claude Sonnet 4.6anthropic57.7%
46Google: Gemma 4 26B A4B (free)google55.7%
47OpenAI: o4 Miniopenai55.0%
48Tencent: Hy3 previewtencent54.7%
49StepFun: Step 3.5 Flashstepfun54.3%
50OpenAI: gpt-oss-120bopenai50.7%
51Google: Gemini 3 Flash Previewgoogle48.0%
52Google: Gemini 2.5 Flash Lite Preview 09-2025google48.0%
53Meta: Llama 4 Maverickmeta-llama46.0%
54Google: Gemini 2.5 Flashgoogle45.9%
55OpenAI: GPT-4.1 Miniopenai42.3%
56OpenAI: GPT-5 Nanoopenai41.7%
57Qwen: Qwen3 Coder Nextqwen40.0%
58OpenAI: o3 Mini Highopenai39.3%
59Inception: Mercury 2inception36.3%
60OpenAI: GPT-4o (2024-08-06)openai35.0%
61inclusionAI: Ling-2.6-1Tinclusionai34.7%
62Arcee AI: Trinity Large Thinkingarcee-ai33.0%
63Prime Intellect: INTELLECT-3prime-intellect32.3%
64Xiaomi: MiMo-V2-Flashxiaomi31.3%
65OpenAI: gpt-oss-20bopenai30.7%
66Upstage: Solar Pro 3upstage27.0%
67Meta: Llama 4 Scoutmeta-llama25.8%
68inclusionAI: Ling-2.6-flashinclusionai25.0%
69Cohere: Command Acohere18.0%
70OpenAI: GPT-4.1 Nanoopenai17.0%
71Microsoft: Phi 4 Mini Instructmicrosoft13.7%
72IBM: Granite 4.1 8Bibm-granite12.0%
73Google: Gemma 3 12Bgoogle6.7%
74Google: Gemma 3 27Bgoogle5.7%
75Google: Gemma 3 4Bgoogle5.7%
76Microsoft: Phi 4microsoft0.0%
77OpenAI: GPT-4oopenai0.0%
78Reka Flash 3rekaai0.0%

Artificial Analysis (artificialanalysis.ai). Redistribution requires an AA commercial license.

← All rankingsMethodology & 𝕍 →