Preservation

Subcategory of Text Editing. 89 models scored.

Model Leaderboard

All models ranked by their Preservation subcategory score.

# Model Preservation Text Editing Overall
1 Claude Opus 4.6 100.00% 98.35% 92.35%
2 Claude Opus 4.5 100.00% 97.69% 89.69%
3 Claude Sonnet 4.6 99.82% 96.37% 91.15%
4 Claude Sonnet 4 99.69% 99.13% 88.72%
5 Claude Haiku 4.5 99.64% 96.81% 85.14%
6 Qwen 3.5 Plus (2026-02-15) 99.38% 98.10% 85.96%
7 Claude Sonnet 4.5 99.31% 99.02% 88.03%
8 Grok 4 99.23% 98.76% 88.12%
9 Claude Sonnet 4.6 (Reasoning) 99.06% 98.30% 93.66%
10 Gemini 3.1 Pro (Preview) 99.03% 98.51% 94.37%
11 Gemini 3 Flash (Preview) 98.90% 97.54% 85.35%
12 Gemini 2.5 Pro 98.80% 98.58% 88.53%
13 Claude Opus 4.6 (Reasoning) 98.72% 98.86% 95.02%
14 Z.AI GLM 4.7 98.67% 98.22% 88.69%
15 Gemini 2.5 Flash 98.67% 97.83% 80.60%
16 Z.AI GLM 5 98.62% 98.59% 91.23%
17 Grok 4 Fast 98.60% 97.26% 86.15%
18 GPT-5 98.57% 98.90% 91.93%
19 Qwen 3.5 27B 98.55% 98.69% 90.85%
20 Gemini 3 Pro (Preview) 98.47% 98.86% 88.79%
21 Claude Opus 4 98.34% 97.25% 87.69%
22 Z.AI GLM 4.6 98.31% 97.78% 89.11%
23 Qwen 3.5 122B 98.28% 96.31% 91.53%
24 MoonshotAI: Kimi K2.5 98.19% 97.79% 91.04%
25 Gemini 3 Flash (Preview, Reasoning) 98.19% 98.12% 90.50%
26 GPT-5.1 98.18% 98.54% 92.54%
27 GPT-4o, May 13th (temp=0) 98.16% 95.35% 85.36%
28 Minimax M2.5 98.01% 96.02% 88.71%
29 Claude 3.7 Sonnet 97.73% 97.12% 83.39%
30 GPT-4.1 Mini 97.60% 95.62% 83.20%
31 ByteDance Seed 1.6 97.58% 98.40% 90.70%
32 GPT-4o, May 13th (temp=1) 97.54% 92.41% 83.80%
33 GPT-5 Mini 97.51% 97.13% 92.62%
34 Hermes 3 405B 97.48% 89.14% 82.86%
35 Gemini 2.5 Flash (Reasoning) 97.39% 98.12% 86.51%
36 Grok 4.1 Fast 97.29% 97.87% 89.55%
37 Aion 2.0 97.21% 95.34% 89.21%
38 o4 Mini High 97.09% 94.36% 90.29%
39 Mistral Small Creative 96.98% 90.31% 73.27%
40 Claude 3.5 Sonnet 96.96% 96.57% 84.24%
41 GPT-4o, Aug. 6th (temp=0) 96.79% 93.77% 82.45%
42 GPT-4.1 96.66% 94.40% 88.68%
43 Mistral Medium 3.1 96.56% 93.77% 77.83%
44 GPT-5.2 96.38% 97.54% 90.26%
45 Qwen 3.5 397B A17B 96.38% 98.05% 91.73%
46 o4 Mini 96.24% 90.61% 88.35%
47 Qwen 2.5 72B 95.78% 89.18% 75.46%
48 Gemma 3 27B 95.49% 86.63% 77.85%
49 ByteDance Seed 1.6 Flash 95.32% 91.64% 73.27%
50 GPT-4o, Aug. 6th (temp=1) 95.01% 86.72% 82.62%
51 Llama 3.1 70B 94.67% 92.10% 78.40%
52 Mistral Large 94.67% 95.14% 80.15%
53 Gemma 3 4B 94.35% 78.38% 68.57%
54 Writer: Palmyra X5 94.28% 91.20% 79.57%
55 Qwen 3.5 35B 94.21% 94.95% 88.00%
56 Llama 3.1 Nemotron 70B 93.45% 87.26% 74.70%
57 Qwen 3.5 Flash 93.04% 92.80% 86.38%
58 DeepSeek V3 (2024-12-26) 92.60% 93.58% 83.68%
59 Mistral Small 3.2 24B 92.38% 89.48% 78.60%
60 Gemini 2.5 Flash Lite 92.17% 92.13% 81.08%
61 GPT-5 Nano 91.91% 82.74% 82.60%
62 DeepSeek V3.2 91.71% 95.78% 82.25%
63 Z.AI GLM 4.5 91.69% 95.32% 86.27%
64 Mistral Large 3 91.61% 94.09% 85.43%
65 Mistral Large 2 91.61% 94.16% 82.41%
66 DeepSeek-V2 Chat 91.53% 90.90% 84.83%
67 DeepSeek V3 (2025-03-24) 90.56% 89.57% 81.99%
68 Z.AI GLM 4.7 Flash 90.40% 85.82% 84.82%
69 WizardLM 2 8x22b 89.67% 88.13% 71.07%
70 Ministral 3 14B 89.12% 86.20% 72.54%
71 Gemini 2.5 Flash Lite (Reasoning) 89.04% 94.54% 85.75%
72 Gemma 3 12B 87.66% 85.18% 78.41%
73 GPT-4o Mini (temp=1) 86.94% 85.78% 79.08%
74 GPT-4o Mini (temp=0) 86.87% 84.62% 78.29%
75 Arcee AI: Trinity Mini 85.07% 73.88% 70.90%
76 DeepSeek V3.1 84.70% 87.27% 82.39%
77 Ministral 8B 83.69% 77.52% 64.87%
78 Ministral 3 8B 83.53% 78.52% 71.76%
79 Arcee AI: Trinity Large (Preview) 83.17% 86.62% 73.33%
80 Mistral NeMO 82.35% 73.69% 65.04%
81 Llama 3.1 8B 80.76% 75.45% 63.37%
82 GPT-4.1 Nano 78.02% 76.06% 71.94%
83 Ministral 3B 74.55% 70.91% 61.29%
84 Ministral 3 3B 73.12% 69.80% 67.22%
85 Cohere Command R+ (Aug. 2024) 70.67% 68.40% 69.03%
86 Hermes 3 70B 66.87% 63.34% 72.57%
87 Rocinante 12B 63.96% 56.31% 54.55%
88 Claude 3 Haiku 63.45% 64.36% 71.19%
89 LFM2 24B 57.43% 71.56% 58.77%