---
title: "Text · style controlled: model rankings and source evidence · Undominated.ai"
canonical: https://undominated.ai/rankings/arena-text-style-control/
description: "Text · style controlled ranked by Arena score. Inspect exact model variants, source dates, measurement limits and downloadable evidence."
---

# Text · style controlled: model rankings and source evidence · Undominated.ai

> Text · style controlled ranked by Arena score. Inspect exact model variants, source dates, measurement limits and downloadable evidence.

[All rankings](/rankings/) / Language models

# Text · style controlled

Human preference ratings for the overall category of this Arena task. Scores from different boards are not interchangeable.

**419 entries** · Arena score · Higher is better

Source publication: 2026-10-28. Retrieved 2026-11-05.

Source: [Arena (lmarena-ai)](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset) · [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/).

Ranks use confidence bounds. Several models can share a rank; the rows within a shared rank are displayed by score.

Text · style controlled · Arena score

| Significance rank | Model / evaluated variant | Arena score | 95% confidence interval | Votes |
| --- | --- | --- | --- | --- |
| 1 | Gemini 4 Carbon (High) google | 1,527.51 | 1,518.1 – 1,536.92 | 4,403 |
| 1 | Gemini 4 Argon (High) google | 1,520.25 | 1,513.98 – 1,526.53 | 11,889 |
| 2 | Claude Opus 5.5 (High) anthropic | 1,508.7 | 1,502.09 – 1,515.3 | 11,415 |
| 2 | Muse Wonder (Max) meta | 1,507.1 | 1,499.76 – 1,514.44 | 7,749 |
| 3 | Claude Fable 5.1 (Max) anthropic | 1,504.37 | 1,498.55 – 1,510.18 | 17,178 |
| 3 | Claude Fable 5 (High) anthropic | 1,503.89 | 1,499.72 – 1,508.07 | 45,341 |
| 3 | Claude Opus 4.6 (High) anthropic | 1,503.16 | 1,499.9 – 1,506.43 | 84,512 |
| 3 | Claude Opus 4.7 (High) anthropic | 1,501.05 | 1,497.29 – 1,504.81 | 65,110 |
| 7 | Claude Opus 4.6 anthropic | 1,496.1 | 1,492.91 – 1,499.28 | 89,324 |
| 4 | Gemini 3.8 Flash (High) google | 1,495.95 | 1,491.51 – 1,500.38 | 47,803 |
| 4 | Muse Spark 1.3 (Max) meta | 1,495.05 | 1,489.8 – 1,500.29 | 22,134 |
| 8 | Claude Opus 4.7 anthropic | 1,493.72 | 1,489.96 – 1,497.48 | 66,140 |
| 8 | Muse Spark 1.2 (xHigh) meta | 1,492.19 | 1,485.84 – 1,498.54 | 11,724 |
| 9 | Claude Opus 5 (High) anthropic | 1,491.55 | 1,487.97 – 1,495.13 | 73,167 |
| 9 | Claude Opus 5 (Max) anthropic | 1,490.5 | 1,485.81 – 1,495.2 | 35,531 |
| 9 | Muse Spark 1.1 meta | 1,490.24 | 1,486.09 – 1,494.38 | 46,354 |
| 9 | Gemini 3.7 Flash (High) google | 1,489.55 | 1,484.88 – 1,494.22 | 32,289 |
| 9 | Muse Spark meta | 1,489.03 | 1,483.3 – 1,494.75 | 14,140 |
| 10 | Kimi K3 (Max) moonshot | 1,488.19 | 1,483.75 – 1,492.63 | 36,780 |
| 12 | Gemini 3.1 Pro Preview google | 1,486.9 | 1,483.98 – 1,489.82 | 132,547 |
| 13 | Gemini 3 Pro google | 1,485.43 | 1,481.64 – 1,489.22 | 42,057 |
| 13 | GPT-5.6 Sol (xHigh) openai | 1,484.91 | 1,480.67 – 1,489.15 | 43,796 |
| 14 | Gemini 3.6 Flash (High) google | 1,483.16 | 1,478.73 – 1,487.58 | 36,028 |
| 14 | Qwen3.8-Max alibaba | 1,482.41 | 1,477.62 – 1,487.2 | 31,915 |
| 17 | Claude Opus 4.8 (High) anthropic | 1,481.71 | 1,477.85 – 1,485.58 | 63,994 |
| 17 | GPT-5.5 (High) openai | 1,481.7 | 1,477.97 – 1,485.42 | 69,377 |
| 13 | GPT-6.1 Sol (Max) openai | 1,481.33 | 1,474.28 – 1,488.38 | 9,367 |
| 17 | GLM-5.3 (Max) zai | 1,480.57 | 1,475.62 – 1,485.52 | 26,606 |
| 13 | MiMo-V2.6-Pro xiaomi | 1,479.7 | 1,470.28 – 1,489.12 | 4,037 |
| 22 | Gemini 3.5 Flash (High) google | 1,477.63 | 1,473.74 – 1,481.52 | 48,376 |
| 19 | Claude Sonnet 5.5 (xHigh) anthropic | 1,477.15 | 1,470.37 – 1,483.93 | 10,033 |
| 23 | GPT-5.5 openai | 1,476.65 | 1,472.97 – 1,480.33 | 70,949 |
| 22 | GLM-5.3-Flash zai | 1,476.52 | 1,471.86 – 1,481.17 | 41,642 |
| 23 | Gemini 3.5 Flash (Medium) google | 1,476.41 | 1,472.26 – 1,480.55 | 47,196 |
| 21 | DeepSeek-V4.1-Flash-20260910 (Max) deepseek | 1,476.23 | 1,470.13 – 1,482.33 | 13,657 |
| 23 | GPT-5.2 Chat (20260210) openai | 1,475.66 | 1,471.62 – 1,479.7 | 35,974 |
| 22 | GPT-6 Astra (Max) openai | 1,475.65 | 1,469.72 – 1,481.57 | 15,038 |
| 23 | GLM-5.2 (Max) zai | 1,475.29 | 1,471.21 – 1,479.37 | 53,702 |
| 23 | GPT-5.4 (High) openai | 1,475.27 | 1,471.51 – 1,479.04 | 64,748 |
| 23 | Grok 4.20 Beta 1 xai | 1,474.61 | 1,470.05 – 1,479.18 | 27,889 |
| 24 | Claude Opus 4.8 anthropic | 1,474.44 | 1,470.6 – 1,478.29 | 65,121 |
| 20 | Qwen3.7-Max-Preview alibaba | 1,473.7 | 1,463.97 – 1,483.44 | 3,932 |
| 27 | Claude Opus 4.5 (High, 32K, 20251101) anthropic | 1,473.69 | 1,469.86 – 1,477.52 | 37,484 |
| 24 | GPT-5.5 Instant openai | 1,473.17 | 1,468.16 – 1,478.17 | 27,128 |
| 27 | Gemini 3 Flash google | 1,472.75 | 1,468.41 – 1,477.09 | 31,301 |
| 28 | Claude Sonnet 4.6 anthropic | 1,471.95 | 1,468.44 – 1,475.46 | 70,916 |
| 28 | Grok 4.20 Beta (Reasoning, 0309) xai | 1,471.74 | 1,468.07 – 1,475.41 | 66,422 |
| 28 | Grok 4.20 Multi-Agent Beta (0309) xai | 1,470.71 | 1,467.01 – 1,474.41 | 64,917 |
| 31 | Claude Opus 4.5 (20251101) anthropic | 1,469.68 | 1,466.54 – 1,472.81 | 72,825 |
| 35 | GPT-5.6 Terra (xHigh) openai | 1,467.34 | 1,463.19 – 1,471.48 | 48,166 |
| 34 | ERNIE 5.1 baidu | 1,467.08 | 1,462.55 – 1,471.62 | 39,469 |
| 40 | MiMo-V2.5-Pro xiaomi | 1,466.6 | 1,463.09 – 1,470.11 | 77,728 |
| 42 | Grok 4.5 xai | 1,465.45 | 1,461.11 – 1,469.78 | 39,801 |
| 43 | GLM-5.1 zai | 1,465.38 | 1,461.86 – 1,468.9 | 66,771 |
| 33 | DeepSeek-V4-Pro (High, 20260813) deepseek | 1,465.06 | 1,458.44 – 1,471.67 | 9,787 |
| 45 | Grok 4.1 (Thinking) xai | 1,465.05 | 1,461.87 – 1,468.23 | 66,470 |
| 43 | GPT-5.4 openai | 1,465.04 | 1,461.36 – 1,468.73 | 67,856 |
| 43 | Qwen3.5-Max-Preview alibaba | 1,464.64 | 1,459.76 – 1,469.52 | 22,582 |
| 49 | Claude Sonnet 5 (High) anthropic | 1,461.74 | 1,457.71 – 1,465.78 | 52,952 |
| 49 | Kimi K2.6 moonshot | 1,460.96 | 1,456.53 – 1,465.38 | 39,978 |
| 44 | Qwen3.6-Max-Preview alibaba | 1,460.21 | 1,452.02 – 1,468.41 | 5,419 |
| 53 | Grok 4.1 xai | 1,459 | 1,455.84 – 1,462.17 | 68,599 |
| 55 | Gemini 3 Flash (Minimal) google | 1,458.37 | 1,455.35 – 1,461.39 | 90,997 |
| 49 | GPT-6 Sol (Max) openai | 1,458.32 | 1,452.44 – 1,464.2 | 15,807 |
| 55 | DeepSeek-V4-Pro deepseek | 1,457.77 | 1,453.88 – 1,461.67 | 57,572 |
| 55 | GLM-5 zai | 1,457.24 | 1,452.99 – 1,461.48 | 29,255 |
| 58 | Claude Sonnet 4.5 (High, 32K, 20250929) anthropic | 1,456.32 | 1,453.57 – 1,459.08 | 83,892 |
| 58 | Dola Seed 2.0 Pro bytedance | 1,456.13 | 1,452.74 – 1,459.52 | 79,094 |
| 58 | GPT-5.1 (High) openai | 1,455.7 | 1,451.99 – 1,459.41 | 41,639 |
| 57 | Qwen3.7-Plus alibaba | 1,455.66 | 1,451.33 – 1,460 | 42,011 |
| 57 | Gemini 3.5 Flash-Lite google | 1,455.66 | 1,451.18 – 1,460.13 | 35,740 |
| 57 | Grok 4.6 (High) xai | 1,455.48 | 1,450.86 – 1,460.1 | 33,824 |
| 59 | Claude Sonnet 4.5 (20250929) anthropic | 1,455.05 | 1,452.19 – 1,457.91 | 82,621 |
| 58 | DeepSeek-V4-Pro Preview (High) deepseek | 1,454.89 | 1,450.93 – 1,458.86 | 54,870 |
| 57 | Hy3 tencent | 1,454.48 | 1,448.6 – 1,460.36 | 13,840 |
| 60 | GPT-5.6 Luna (xHigh) openai | 1,453.02 | 1,448.72 – 1,457.32 | 41,317 |
| 58 | Gemma 4 31B google | 1,452.29 | 1,444.85 – 1,459.73 | 6,138 |
| 63 | GPT-5.3 Chat openai | 1,450.11 | 1,445.87 – 1,454.34 | 34,316 |
| 65 | Claude Opus 4.1 (Thinking 16K, 20250805) anthropic | 1,449.77 | 1,446.32 – 1,453.22 | 49,535 |
| 65 | Kimi K2.5 (Thinking) moonshot | 1,449.75 | 1,446.47 – 1,453.02 | 74,682 |
| 63 | ERNIE-5.0-Preview (1203) baidu | 1,448.49 | 1,441.99 – 1,454.98 | 9,802 |
| 63 | Step 5 Preview stepfun | 1,448.37 | 1,442 – 1,454.75 | 11,355 |
| 73 | Claude Opus 4.1 (20250805) anthropic | 1,448.01 | 1,445.04 – 1,450.98 | 77,213 |
| 68 | MiMo-V2-Pro xiaomi | 1,447.76 | 1,443.1 – 1,452.42 | 25,421 |
| 65 | MiMo-V2.6-Flash xiaomi | 1,447.59 | 1,441.79 – 1,453.39 | 15,312 |
| 74 | GPT-5.4 mini (High) openai | 1,447.17 | 1,443.43 – 1,450.92 | 63,432 |
| 75 | ERNIE 5.0 (0110) baidu | 1,446.14 | 1,442.25 – 1,450.04 | 36,475 |
| 77 | Gemini 2.5 Pro google | 1,445.41 | 1,442.95 – 1,447.87 | 125,307 |
| 75 | GPT-4.5 Preview (2025-02-27) openai | 1,444.64 | 1,438.92 – 1,450.35 | 14,547 |
| 77 | Qwen3.6-Plus alibaba | 1,443 | 1,438.87 – 1,447.13 | 48,126 |
| 80 | ChatGPT-4o Latest (20250326) openai | 1,442.65 | 1,439.84 – 1,445.46 | 81,893 |
| 78 | Inkling thinky | 1,442.2 | 1,437.99 – 1,446.41 | 45,472 |
| 80 | Grok 4.3 xai | 1,441.63 | 1,438 – 1,445.26 | 71,287 |
| 82 | Qwen3.5-397B-A17B alibaba | 1,441.4 | 1,438.35 – 1,444.46 | 97,675 |
| 77 | GLM-4.7 zai | 1,441.34 | 1,435.29 – 1,447.4 | 12,239 |
| 77 | Grok 4.7 (xHigh) xai | 1,441.23 | 1,435.48 – 1,446.97 | 16,442 |
| 80 | GPT-6 Luna (Max) openai | 1,440.87 | 1,436.59 – 1,445.14 | 86,478 |
| 84 | MiniMax-M3 minimax | 1,439.24 | 1,435.49 – 1,442.99 | 68,611 |
| 85 | DeepSeek-V4-Flash Preview (High) deepseek | 1,438.65 | 1,434.55 – 1,442.76 | 51,763 |
| 86 | GPT-5.1 openai | 1,438.53 | 1,434.94 – 1,442.12 | 44,413 |
| 81 | Gemma 4 26B A4B google | 1,437.6 | 1,430.16 – 1,445.04 | 6,090 |
| 89 | GPT-5.2 (High) openai | 1,437.29 | 1,433.74 – 1,440.84 | 49,628 |
| 89 | LongCat-Flash-Chat Exp (2602) meituan | 1,436.4 | 1,431.87 – 1,440.93 | 29,244 |
| 90 | DeepSeek-V4-Flash deepseek | 1,435.67 | 1,431.64 – 1,439.71 | 51,944 |
| 81 | Beam 1.0 501B A23B reflection | 1,435.67 | 1,426.35 – 1,444.99 | 4,233 |
| 89 | Qwen3.8-27B alibaba | 1,435.6 | 1,430.73 – 1,440.47 | 27,779 |
| 92 | GPT-5.2 openai | 1,435.55 | 1,432.38 – 1,438.73 | 83,793 |
| 90 | GPT-5 (High) openai | 1,434.74 | 1,430.24 – 1,439.25 | 31,718 |
| 92 | Qwen3-Max-Preview alibaba | 1,434.26 | 1,429.75 – 1,438.78 | 27,368 |
| 89 | GLM-5V-Turbo zai | 1,433.96 | 1,427.52 – 1,440.39 | 10,476 |
| 93 | MiMo-V2.5 xiaomi | 1,433.91 | 1,429.69 – 1,438.13 | 47,327 |
| 96 | Gemini 3.1 Flash-Lite Preview google | 1,432.9 | 1,429.28 – 1,436.52 | 64,196 |
| 93 | Mistral Large 4 mistral | 1,432.01 | 1,425.8 – 1,438.23 | 12,066 |
| 98 | o3 (2025-04-16) openai | 1,431.66 | 1,428.03 – 1,435.29 | 58,901 |
| 96 | MiMo-V2-Omni xiaomi | 1,430.73 | 1,425.19 – 1,436.28 | 20,615 |
| 95 | Kimi K2.5 (Instant) moonshot | 1,430.52 | 1,424.02 – 1,437.03 | 8,371 |
| 102 | Grok 4.1 Fast (Reasoning) xai | 1,430.3 | 1,427.11 – 1,433.5 | 57,500 |
| 102 | Kimi K2 Thinking Turbo moonshot | 1,429.97 | 1,426.88 – 1,433.06 | 63,310 |
| 96 | amazon-nova-experimental-chat (26-02-10) amazon | 1,426.71 | 1,417.04 – 1,436.38 | 3,523 |
| 102 | NVIDIA-Nemotron-3-Ultra-550B-A55B (NVFP4) nvidia | 1,426.51 | 1,419.62 – 1,433.4 | 11,563 |
| 105 | GPT-5 Chat openai | 1,426.44 | 1,422.15 – 1,430.74 | 31,187 |
| 102 | Mistral Medium 3.5 mistral | 1,426.38 | 1,419.99 – 1,432.77 | 11,791 |
| 106 | Claude Opus 4 (Thinking 16K, 20250514) anthropic | 1,426.23 | 1,421.83 – 1,430.63 | 35,904 |
| 104 | DeepSeek-V3.2-Exp (Thinking) deepseek | 1,425.15 | 1,418.58 – 1,431.71 | 8,991 |
| 100 | Muse Glimmer meta | 1,424.94 | 1,415.23 – 1,434.64 | 3,884 |
| 103 | Claude Haiku 5.5 (High) anthropic | 1,424.82 | 1,417.34 – 1,432.31 | 7,890 |
| 111 | DeepSeek-V3.2 deepseek | 1,424.65 | 1,421.16 – 1,428.14 | 48,095 |
| 111 | GLM-4.6 zai | 1,424.29 | 1,420.39 – 1,428.19 | 35,810 |
| 108 | Qwen3-Max (2025-09-23) alibaba | 1,423.49 | 1,417.01 – 1,429.98 | 9,026 |
| 116 | DeepSeek-V3.2 (Thinking) deepseek | 1,422.34 | 1,418.77 – 1,425.91 | 42,033 |
| 118 | Qwen3-235B-A22B-Instruct (2507) alibaba | 1,422.21 | 1,419.66 – 1,424.76 | 98,081 |
| 111 | DeepSeek-V3.2-Exp deepseek | 1,421.72 | 1,415.34 – 1,428.09 | 11,996 |
| 113 | DeepSeek-R1 (0528) deepseek | 1,421.46 | 1,415.78 – 1,427.14 | 18,165 |
| 118 | Kimi K2 Preview (0905) moonshot | 1,418.47 | 1,411.94 – 1,425 | 11,592 |
| 116 | Grok 4 Fast (Non-Reasoning) xai | 1,418.4 | 1,410.6 – 1,426.21 | 6,336 |
| 114 | ERNIE-5.0-Preview (1022) baidu | 1,418.27 | 1,409.51 – 1,427.02 | 4,755 |
| 119 | Kimi K2 Preview (0711) moonshot | 1,417.78 | 1,412.88 – 1,422.67 | 27,139 |
| 119 | DeepSeek-V3.1 deepseek | 1,416.94 | 1,410.88 – 1,423.01 | 14,606 |
| 122 | Qwen3.5-122B-A10B alibaba | 1,416.46 | 1,412.17 – 1,420.74 | 29,712 |
| 116 | DeepSeek-V3.1-Terminus (Thinking) deepseek | 1,416.1 | 1,405.99 – 1,426.2 | 3,379 |
| 119 | DeepSeek-V3.1 (Thinking) deepseek | 1,415.8 | 1,409.14 – 1,422.46 | 11,458 |
| 128 | MiniMax-M2.7 minimax | 1,414.99 | 1,411.68 – 1,418.29 | 89,197 |
| 118 | DeepSeek-V3.1-Terminus deepseek | 1,414.71 | 1,404.99 – 1,424.43 | 3,609 |
| 128 | GPT-4.1 (2025-04-14) openai | 1,414.64 | 1,410.93 – 1,418.35 | 50,298 |
| 131 | Claude Haiku 4.5 (20251001) anthropic | 1,414.5 | 1,412.12 – 1,416.89 | 154,063 |
| 129 | Mistral Large 3 mistral | 1,414.35 | 1,411.53 – 1,417.17 | 88,325 |
| 127 | Claude Opus 4 (20250514) anthropic | 1,414.31 | 1,409.99 – 1,418.63 | 43,037 |
| 124 | Qwen3-VL-235B-A22B-Instruct alibaba | 1,413.29 | 1,406.76 – 1,419.81 | 11,314 |
| 119 | amazon-nova-experimental-chat (26-01-10) amazon | 1,412.76 | 1,403.09 – 1,422.44 | 3,534 |
| 121 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4) nvidia | 1,412.27 | 1,402.91 – 1,421.64 | 4,178 |
| 132 | Grok 3 Preview (02-24) xai | 1,411.21 | 1,406.87 – 1,415.56 | 32,496 |
| 134 | Grok 4 (0709) xai | 1,410.98 | 1,407.04 – 1,414.93 | 39,678 |
| 131 | GLM-4.5 zai | 1,410.88 | 1,405.96 – 1,415.8 | 23,797 |
| 139 | Gemini 2.5 Flash google | 1,409.24 | 1,406.8 – 1,411.67 | 125,457 |
| 134 | Qwen3.5-27B alibaba | 1,408.81 | 1,404.51 – 1,413.1 | 28,647 |
| 131 | Hy3 preview tencent | 1,408.76 | 1,401.35 – 1,416.17 | 6,869 |
| 143 | Mistral Medium 3.1 (2508) mistral | 1,407.79 | 1,405.2 – 1,410.39 | 95,269 |
| 143 | Inkling-Small thinky | 1,405.52 | 1,401.01 – 1,410.04 | 36,999 |
| 143 | Grok 4 Fast (Reasoning) xai | 1,405.28 | 1,400.22 – 1,410.35 | 18,270 |
| 146 | Gemini 2.5 Flash Preview (09-2025) google | 1,403.04 | 1,398.99 – 1,407.09 | 32,996 |
| 150 | o1 (2024-12-17) openai | 1,402.27 | 1,397.79 – 1,406.74 | 27,807 |
| 150 | Qwen3-235B-A22B (Non-Thinking) alibaba | 1,402.12 | 1,397.58 – 1,406.65 | 37,472 |
| 150 | Claude Sonnet 4 (Thinking 32K, 20250514) anthropic | 1,401.95 | 1,397.5 – 1,406.4 | 34,128 |
| 154 | GPT-5.4 nano (High) openai | 1,401.19 | 1,397.41 – 1,404.97 | 62,314 |
| 146 | LongCat-Flash-Chat meituan | 1,400.77 | 1,394.36 – 1,407.19 | 11,167 |
| 150 | Qwen3-235B-A22B-Thinking (2507) alibaba | 1,400.06 | 1,393.48 – 1,406.63 | 8,822 |
| 155 | Qwen3-Next-80B-A3B-Instruct alibaba | 1,398.51 | 1,393.74 – 1,403.29 | 22,753 |
| 156 | DeepSeek-R1 deepseek | 1,398.08 | 1,393.18 – 1,402.98 | 18,524 |
| 160 | Qwen3.5-Flash alibaba | 1,396 | 1,392.33 – 1,399.66 | 61,776 |
| 160 | DeepSeek-V3 (0324) deepseek | 1,395.24 | 1,391.34 – 1,399.13 | 44,977 |
| 157 | Qwen3-VL-235B-A22B-Thinking alibaba | 1,394.71 | 1,387.89 – 1,401.54 | 7,862 |
| 150 | Hunyuan Vision 1.5 Thinking tencent | 1,394.07 | 1,381.66 – 1,406.48 | 2,169 |
| 161 | Qwen3.5-35B-A3B alibaba | 1,393.83 | 1,389.6 – 1,398.06 | 30,427 |
| 165 | Step-3.5-Flash stepfun | 1,393.01 | 1,389.46 – 1,396.56 | 60,305 |
| 157 | amazon-nova-experimental-chat (12-10) amazon | 1,392.91 | 1,383.44 – 1,402.37 | 3,734 |
| 165 | MiMo-V2-Flash (Non-Thinking) xiaomi | 1,391.51 | 1,388.03 – 1,394.99 | 48,060 |
| 165 | o4-mini (2025-04-16) openai | 1,390.95 | 1,386.96 – 1,394.95 | 44,843 |
| 165 | Claude Sonnet 4 (20250514) anthropic | 1,390.93 | 1,386.53 – 1,395.32 | 39,168 |
| 165 | MiniMax-M2.5 minimax | 1,390.62 | 1,386.74 – 1,394.51 | 42,783 |
| 165 | GPT-5 mini (High) openai | 1,389.83 | 1,385.19 – 1,394.48 | 26,793 |
| 167 | o1-preview openai | 1,388.57 | 1,383.56 – 1,393.58 | 31,122 |
| 169 | Claude 3.7 Sonnet (Thinking 32K, 20250219) anthropic | 1,388.26 | 1,384.05 – 1,392.46 | 38,156 |
| 169 | Solar Pro 4 upstage | 1,387.83 | 1,382.78 – 1,392.87 | 26,159 |
| 169 | Qwen3-Coder-480B-A35B-Instruct alibaba | 1,387.57 | 1,382.59 – 1,392.55 | 25,190 |
| 165 | Hunyuan T1 (20250711) tencent | 1,387.27 | 1,378.56 – 1,395.98 | 4,595 |
| 170 | Mistral Medium 3 (2505) mistral | 1,387.26 | 1,382.54 – 1,391.98 | 32,777 |
| 170 | MiMo-V2-Flash (Thinking) xiaomi | 1,386.11 | 1,379.94 – 1,392.28 | 11,135 |
| 171 | MiniMax-M2.1 Preview minimax | 1,384.54 | 1,379.36 – 1,389.72 | 17,222 |
| 172 | Hunyuan TurboS (20250416) tencent | 1,383.09 | 1,376.59 – 1,389.59 | 10,623 |
| 175 | GPT-4.1 mini (2025-04-14) openai | 1,382.81 | 1,378.48 – 1,387.15 | 38,696 |
| 175 | Qwen3-30B-A3B-Instruct (2507) alibaba | 1,382.31 | 1,377.4 – 1,387.22 | 23,312 |
| 182 | Gemini 2.5 Flash-Lite Preview (Non-Thinking, 09-2025) google | 1,379.4 | 1,375.97 – 1,382.82 | 47,418 |
| 182 | Trinity Large Preview Organization not supplied | 1,378.53 | 1,374.27 – 1,382.79 | 31,233 |
| 172 | GLM-4.6V zai | 1,378.36 | 1,367.18 – 1,389.54 | 2,833 |
| 187 | Qwen3-235B-A22B alibaba | 1,374.85 | 1,370.15 – 1,379.56 | 25,950 |
| 188 | Gemini 2.5 Flash-Lite Preview (Thinking, 06-17) google | 1,374.7 | 1,370.17 – 1,379.22 | 32,381 |
| 190 | Claude 3.5 Sonnet (20241022) anthropic | 1,374.28 | 1,371.07 – 1,377.49 | 87,771 |
| 190 | Qwen2.5-Max alibaba | 1,373.77 | 1,369.65 – 1,377.88 | 32,487 |
| 191 | GLM-4.5-Air zai | 1,373.02 | 1,368.77 – 1,377.28 | 30,652 |
| 192 | Claude 3.7 Sonnet (20250219) anthropic | 1,372.47 | 1,368.5 – 1,376.44 | 42,573 |
| 193 | Qwen3-Next-80B-A3B-Thinking alibaba | 1,369.19 | 1,363.33 – 1,375.05 | 13,476 |
| 194 | Trinity Large Thinking Organization not supplied | 1,367.48 | 1,362.89 – 1,372.08 | 30,290 |
| 198 | Gemma 3 27B IT google | 1,365.32 | 1,361.62 – 1,369.03 | 46,670 |
| 195 | GLM-4.7-Flash zai | 1,364.71 | 1,358.99 – 1,370.42 | 12,008 |
| 200 | MiniMax-M1 minimax | 1,363.82 | 1,359.52 – 1,368.11 | 34,293 |
| 200 | amazon-nova-experimental-chat (11-10) amazon | 1,363.81 | 1,359.53 – 1,368.09 | 26,026 |
| 198 | Grok 3 Mini (High) xai | 1,363.7 | 1,358.23 – 1,369.18 | 15,595 |
| 198 | o3-mini (High) openai | 1,363.56 | 1,358.29 – 1,368.83 | 18,589 |
| 201 | Gemini 2.0 Flash (001) google | 1,359.97 | 1,356.16 – 1,363.77 | 43,443 |
| 201 | NVIDIA-Nemotron-3-Super-120B-A12B nvidia | 1,359.93 | 1,352.8 – 1,367.05 | 7,870 |
| 202 | DeepSeek-V3 deepseek | 1,358.42 | 1,353.66 – 1,363.18 | 21,770 |
| 202 | Grok 3 Mini Beta xai | 1,357.89 | 1,352.76 – 1,363.03 | 20,930 |
| 204 | Mistral Small 3.2 (2506) mistral | 1,356.17 | 1,350.95 – 1,361.4 | 17,462 |
| 201 | INTELLECT-3 Organization not supplied | 1,355.51 | 1,347.26 – 1,363.76 | 5,435 |
| 209 | Command A (03-2025) cohere | 1,353.65 | 1,350.18 – 1,357.13 | 55,763 |
| 209 | Gemini 2.0 Flash-Lite Preview (02-05) google | 1,353.54 | 1,349.22 – 1,357.87 | 24,955 |
| 207 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B (NVFP4) nvidia | 1,353.31 | 1,348 – 1,358.62 | 24,230 |
| 204 | GLM-4.5V zai | 1,352.15 | 1,343.71 – 1,360.59 | 4,817 |
| 210 | gpt-oss-120b openai | 1,351.74 | 1,347.37 – 1,356.1 | 30,200 |
| 210 | Gemini 1.5 Pro (002) google | 1,351.3 | 1,347.9 – 1,354.7 | 55,606 |
| 209 | Step-3 stepfun | 1,349.03 | 1,341.52 – 1,356.55 | 6,291 |
| 204 | Hunyuan TurboS (20250226) tencent | 1,348.84 | 1,337.13 – 1,360.54 | 2,220 |
| 210 | amazon-nova-experimental-chat (10-20) amazon | 1,348.2 | 1,342.06 – 1,354.34 | 11,625 |
| 213 | o3-mini openai | 1,348.06 | 1,344.5 – 1,351.61 | 56,735 |
| 206 | Llama 3.1 Nemotron Ultra 253B v1 nvidia | 1,347.57 | 1,335.9 – 1,359.25 | 2,549 |
| 209 | Qwen3-32B alibaba | 1,346.82 | 1,337.36 – 1,356.28 | 3,926 |
| 209 | amazon-nova-experimental-chat (10-09) amazon | 1,346.2 | 1,335.27 – 1,357.13 | 2,781 |
| 215 | GPT-4o (2024-05-13) openai | 1,346.12 | 1,342.66 – 1,349.59 | 112,881 |
| 210 | Qwen-Plus (0125) alibaba | 1,345.89 | 1,337.56 – 1,354.22 | 5,819 |
| 213 | MiniMax-M2 minimax | 1,343.54 | 1,335.79 – 1,351.28 | 6,850 |
| 220 | Claude 3.5 Sonnet (20240620) anthropic | 1,343.28 | 1,339.84 – 1,346.73 | 82,419 |
| 214 | Ling-flash-2.0 ant-group | 1,343.27 | 1,335.99 – 1,350.55 | 6,952 |
| 211 | Llama 3.3 Nemotron Super 49B v1.5 nvidia | 1,343.07 | 1,333.07 – 1,353.06 | 3,327 |
| 211 | Mercury 2 inception-ai | 1,342.82 | 1,332.29 – 1,353.35 | 3,191 |
| 213 | GLM-4-Plus (0111) zai | 1,342.72 | 1,334.3 – 1,351.13 | 5,760 |
| 213 | Gemma 3 12B IT google | 1,341.74 | 1,332.23 – 1,351.25 | 3,829 |
| 213 | Hunyuan Turbo (0110) tencent | 1,340.66 | 1,329.14 – 1,352.18 | 2,290 |
| 214 | Granite 4.2 30B ibm | 1,340.12 | 1,329.63 – 1,350.61 | 3,389 |
| 220 | GPT-5 nano (High) openai | 1,337.65 | 1,330.73 – 1,344.57 | 8,134 |
| 225 | o1-mini openai | 1,336.81 | 1,333.15 – 1,340.48 | 51,981 |
| 225 | Gemini Advanced (0514) google | 1,335.97 | 1,330.81 – 1,341.12 | 50,148 |
| 226 | Grok 2 (2024-08-13) xai | 1,335.65 | 1,331.96 – 1,339.35 | 63,498 |
| 226 | GPT-4o (2024-08-06) openai | 1,335.59 | 1,331.37 – 1,339.8 | 45,499 |
| 225 | QwQ-32B alibaba | 1,335.58 | 1,331.12 – 1,340.04 | 25,177 |
| 226 | Llama 3.1 405B Instruct (BF16) meta | 1,335.45 | 1,331.72 – 1,339.18 | 41,375 |
| 225 | Amazon Nova 2 Lite amazon | 1,334.96 | 1,328.87 – 1,341.04 | 12,415 |
| 228 | Llama 3.1 405B Instruct (FP8) meta | 1,333.6 | 1,330 – 1,337.21 | 59,656 |
| 223 | Step-2 16K Exp (202412) stepfun | 1,333.51 | 1,324.92 – 1,342.1 | 4,833 |
| 232 | Olmo 3.1 32B Instruct allenai | 1,329.32 | 1,323.18 – 1,335.46 | 11,805 |
| 234 | Yi Lightning Organization not supplied | 1,328.28 | 1,323.33 – 1,333.22 | 27,332 |
| 225 | Llama 3.3 Nemotron Super 49B v1 nvidia | 1,327.74 | 1,315.61 – 1,339.87 | 2,218 |
| 241 | Llama 4 Maverick 17B 128E Instruct meta | 1,326.96 | 1,322.68 – 1,331.24 | 39,439 |
| 240 | Qwen3-30B-A3B alibaba | 1,326.7 | 1,321.95 – 1,331.45 | 26,156 |
| 229 | Hunyuan Large (2025-02-10) tencent | 1,326.33 | 1,316.6 – 1,336.05 | 3,738 |
| 248 | Claude 3.5 Haiku (20241022) anthropic | 1,324.76 | 1,321.5 – 1,328.03 | 69,056 |
| 248 | GPT-4 Turbo (2024-04-09) openai | 1,324.5 | 1,320.58 – 1,328.43 | 98,114 |
| 248 | Gemini 1.5 Pro (001) google | 1,324.06 | 1,320.07 – 1,328.05 | 79,138 |
| 240 | DeepSeek-V2.5 (1210) deepseek | 1,323.38 | 1,315.11 – 1,331.65 | 6,795 |
| 245 | Ring-flash-2.0 ant-group | 1,322.36 | 1,315.09 – 1,329.64 | 6,987 |
| 245 | GPT-4.1 nano (2025-04-14) openai | 1,322.26 | 1,314.53 – 1,329.99 | 6,103 |
| 248 | Claude 3 Opus (20240229) anthropic | 1,322.24 | 1,319.17 – 1,325.31 | 194,909 |
| 248 | Llama 4 Scout 17B 16E Instruct meta | 1,321.45 | 1,316.71 – 1,326.18 | 29,823 |
| 223 | Molmo 2 8B allenai | 1,321.35 | 1,300.24 – 1,342.46 | 822 |
| 248 | Step-1o Turbo (202506) stepfun | 1,319.76 | 1,312.88 – 1,326.63 | 8,906 |
| 249 | GLM-4-Plus zai | 1,319.37 | 1,314.43 – 1,324.3 | 26,126 |
| 249 | Qwen-Max (0919) alibaba | 1,318.11 | 1,312.41 – 1,323.82 | 16,478 |
| 253 | Llama 3.3 70B Instruct meta | 1,318.05 | 1,314.54 – 1,321.55 | 54,493 |
| 254 | GPT-4o mini (2024-07-18) openai | 1,317.59 | 1,314.04 – 1,321.15 | 68,705 |
| 251 | Gemma 3n E4B IT google | 1,317.54 | 1,312.36 – 1,322.72 | 22,136 |
| 249 | gpt-oss-20b openai | 1,317.31 | 1,310.89 – 1,323.72 | 10,416 |
| 254 | Qwen2.5-Plus (1127) alibaba | 1,314.61 | 1,308.25 – 1,320.97 | 10,187 |
| 257 | Mistral Large 2 (2407) mistral | 1,314.43 | 1,310.5 – 1,318.37 | 45,459 |
| 257 | Athene V2 Chat Organization not supplied | 1,314.13 | 1,309.57 – 1,318.7 | 24,739 |
| 257 | NVIDIA-Nemotron-3-Nano-30B-A3B (BF16) nvidia | 1,313.61 | 1,308.13 – 1,319.09 | 15,761 |
| 257 | GPT-4 Turbo Preview (0125) openai | 1,313.25 | 1,309.17 – 1,317.34 | 93,439 |
| 257 | GPT-4 Turbo Preview (1106) openai | 1,312.91 | 1,309.02 – 1,316.8 | 100,105 |
| 254 | Hunyuan Standard (2025-02-10) tencent | 1,311.27 | 1,301.61 – 1,320.94 | 3,904 |
| 266 | Gemini 1.5 Flash (002) google | 1,309.27 | 1,305.04 – 1,313.5 | 34,902 |
| 269 | Grok 2 mini (2024-08-13) xai | 1,308.39 | 1,304.66 – 1,312.13 | 52,567 |
| 269 | DeepSeek-V2.5 deepseek | 1,307.18 | 1,302.47 – 1,311.9 | 24,572 |
| 269 | Athene 70B (0725) Organization not supplied | 1,306.63 | 1,300.96 – 1,312.3 | 19,621 |
| 265 | Olmo 3 32B Think allenai | 1,305.95 | 1,297.7 – 1,314.19 | 5,827 |
| 256 | Mercury inception-ai | 1,305.66 | 1,291.9 – 1,319.43 | 1,966 |
| 271 | Mistral Large (2411) mistral | 1,305.56 | 1,301.1 – 1,310.02 | 28,073 |
| 269 | Magistral Medium (2506) mistral | 1,304.82 | 1,298.37 – 1,311.27 | 11,513 |
| 266 | Granite 4.1 8B ibm | 1,303.96 | 1,293.96 – 1,313.96 | 4,177 |
| 267 | Gemma 3 4B IT google | 1,303.23 | 1,293.91 – 1,312.55 | 4,171 |
| 276 | Qwen2.5-72B-Instruct alibaba | 1,302.86 | 1,298.78 – 1,306.95 | 39,406 |
| 276 | Mistral Small 3.1 24B Instruct (2503) mistral | 1,302.59 | 1,298.05 – 1,307.14 | 32,722 |
| 276 | Llama 3.1 Nemotron 70B Instruct nvidia | 1,298.73 | 1,290.97 – 1,306.49 | 7,140 |
| 278 | Hunyuan Large Vision tencent | 1,293.73 | 1,284.59 – 1,302.88 | 5,279 |
| 287 | Llama 3.1 70B Instruct meta | 1,293.35 | 1,289.6 – 1,297.1 | 55,240 |
| 287 | Amazon Nova Pro 1.0 amazon | 1,290.09 | 1,285.51 – 1,294.68 | 24,745 |
| 280 | Granite 4.2 3B ibm | 1,290 | 1,278.4 – 1,301.6 | 3,249 |
| 282 | Granite 4.2 8B ibm | 1,289.68 | 1,278.5 – 1,300.87 | 3,251 |
| 289 | Gemma 2 27B IT google | 1,289.49 | 1,286.09 – 1,292.88 | 75,754 |
| 287 | Jamba 1.5 Large Organization not supplied | 1,289.47 | 1,282.17 – 1,296.77 | 8,662 |
| 287 | Reka Core (20240904) Organization not supplied | 1,288.05 | 1,280.86 – 1,295.23 | 7,312 |
| 289 | GPT-4 (0314) openai | 1,287.89 | 1,283.14 – 1,292.65 | 54,173 |
| 287 | Granite-4.0-H-Small ibm | 1,287.31 | 1,278.91 – 1,295.71 | 5,630 |
| 287 | Llama 3.1 Nemotron 51B Instruct nvidia | 1,286.66 | 1,276.7 – 1,296.63 | 3,749 |
| 290 | Gemini 1.5 Flash (001) google | 1,286.54 | 1,282.04 – 1,291.04 | 62,833 |
| 289 | Olmo 3.1 32B Think allenai | 1,286.44 | 1,279.18 – 1,293.71 | 8,299 |
| 287 | Llama 3.1 Tulu 3 70B allenai | 1,286.11 | 1,275.66 – 1,296.57 | 2,846 |
| 294 | Claude 3 Sonnet (20240229) anthropic | 1,281.4 | 1,277.42 – 1,285.39 | 109,284 |
| 292 | Gemma 2 9B IT SimPO Organization not supplied | 1,280.38 | 1,273.52 – 1,287.24 | 10,072 |
| 296 | Nemotron-4 340B Instruct nvidia | 1,277.2 | 1,271.89 – 1,282.52 | 19,659 |
| 299 | Llama 3 70B Instruct meta | 1,276.6 | 1,273.01 – 1,280.18 | 156,876 |
| 296 | Command R+ (08-2024) cohere | 1,276.34 | 1,269.77 – 1,282.92 | 9,866 |
| 299 | GPT-4 (0613) openai | 1,276.28 | 1,272.23 – 1,280.33 | 88,723 |
| 299 | Mistral Small 24B Instruct (2501) mistral | 1,274.38 | 1,268.46 – 1,280.3 | 14,681 |
| 299 | GLM-4 (0520) zai | 1,273.45 | 1,266.47 – 1,280.43 | 9,788 |
| 299 | Reka Flash (20240904) Organization not supplied | 1,272.45 | 1,265.45 – 1,279.45 | 7,536 |
| 301 | Qwen2.5-Coder-32B-Instruct alibaba | 1,270.63 | 1,262.5 – 1,278.75 | 5,432 |
| 309 | Aya Expanse 32B cohere | 1,267.27 | 1,262.37 – 1,272.17 | 27,124 |
| 310 | Gemma 2 9B IT google | 1,267.02 | 1,263.18 – 1,270.85 | 54,611 |
| 310 | DeepSeek-Coder-V2 deepseek | 1,265.41 | 1,259.14 – 1,271.68 | 15,147 |
| 312 | Qwen2-72B-Instruct alibaba | 1,261.87 | 1,256.92 – 1,266.81 | 37,325 |
| 313 | Command R+ cohere | 1,261.83 | 1,257.52 – 1,266.15 | 77,554 |
| 313 | Claude 3 Haiku (20240307) anthropic | 1,261.73 | 1,257.97 – 1,265.49 | 117,701 |
| 313 | Amazon Nova Lite 1.0 amazon | 1,260.44 | 1,255.24 – 1,265.63 | 19,372 |
| 315 | Gemini 1.5 Flash-8B (001) google | 1,258.8 | 1,254.49 – 1,263.11 | 35,558 |
| 317 | Phi-4 microsoft | 1,256.29 | 1,251.68 – 1,260.9 | 24,126 |
| 316 | OLMo 2 32B Instruct (0325) allenai | 1,251.66 | 1,240.88 – 1,262.43 | 3,334 |
| 320 | Command R (08-2024) cohere | 1,250.48 | 1,243.88 – 1,257.08 | 10,140 |
| 324 | Mistral Large (2402) mistral | 1,242.61 | 1,237.9 – 1,247.32 | 62,436 |
| 324 | Amazon Nova Micro 1.0 amazon | 1,240.71 | 1,235.53 – 1,245.89 | 19,364 |
| 324 | Jamba 1.5 Mini Organization not supplied | 1,239.8 | 1,232.57 – 1,247.02 | 8,858 |
| 324 | Ministral 8B (2410) mistral | 1,237.62 | 1,228.54 – 1,246.71 | 4,781 |
| 324 | Gemini Pro (Dev API) google | 1,236.88 | 1,229.58 – 1,244.19 | 18,354 |
| 326 | Qwen1.5-110B-Chat alibaba | 1,234.22 | 1,228.68 – 1,239.77 | 26,195 |
| 325 | Reka Flash 21B (Online, 20240226) Organization not supplied | 1,233.62 | 1,226.23 – 1,241.01 | 15,450 |
| 326 | Qwen1.5-72B-Chat alibaba | 1,233.59 | 1,228.33 – 1,238.85 | 39,302 |
| 324 | Hunyuan Standard 256K tencent | 1,233.2 | 1,221.48 – 1,244.92 | 2,728 |
| 328 | Mixtral 8x22B Instruct v0.1 mistral | 1,229.65 | 1,225.1 – 1,234.21 | 51,416 |
| 329 | Command R cohere | 1,226.94 | 1,222.16 – 1,231.71 | 54,036 |
| 328 | Reka Flash 21B (20240226) Organization not supplied | 1,226.88 | 1,220.94 – 1,232.81 | 24,806 |
| 329 | GPT-3.5 Turbo (0125) openai | 1,225.64 | 1,220.95 – 1,230.33 | 66,207 |
| 333 | Llama 3 8B Instruct meta | 1,223.68 | 1,219.94 – 1,227.41 | 104,642 |
| 328 | Gemini Pro google | 1,223.67 | 1,211.95 – 1,235.39 | 6,390 |
| 329 | Aya Expanse 8B cohere | 1,223.08 | 1,216.1 – 1,230.06 | 9,818 |
| 333 | Mistral Medium mistral | 1,222.78 | 1,217.3 – 1,228.26 | 34,550 |
| 329 | Llama 3.1 Tulu 3 8B allenai | 1,220.5 | 1,209.83 – 1,231.17 | 2,896 |
| 335 | Zephyr ORPO 141B A35B v0.1 Organization not supplied | 1,213.09 | 1,202.32 – 1,223.86 | 4,652 |
| 340 | Yi 1.5 34B Chat Organization not supplied | 1,212.88 | 1,207.85 – 1,217.91 | 24,146 |
| 342 | Llama 3.1 8B Instruct meta | 1,211.23 | 1,207.07 – 1,215.39 | 49,605 |
| 340 | Granite 3.1 8B Instruct ibm | 1,208.57 | 1,197.49 – 1,219.66 | 3,090 |
| 342 | GPT-3.5 Turbo (1106) openai | 1,204.29 | 1,195.58 – 1,213.01 | 16,619 |
| 343 | Qwen1.5-32B-Chat alibaba | 1,204.06 | 1,197.95 – 1,210.16 | 21,741 |
| 346 | Gemma 2 2B IT google | 1,200.09 | 1,195.99 – 1,204.2 | 46,616 |
| 346 | Phi-3 Medium 4K Instruct microsoft | 1,197.97 | 1,192.81 – 1,203.12 | 25,055 |
| 347 | Mixtral 8x7B Instruct v0.1 mistral | 1,197.23 | 1,192.97 – 1,201.48 | 73,503 |
| 347 | DBRX Instruct Preview Organization not supplied | 1,195.58 | 1,189.48 – 1,201.69 | 32,191 |
| 347 | Qwen1.5-14B-Chat alibaba | 1,191.14 | 1,184.04 – 1,198.24 | 17,839 |
| 347 | InternLM2.5 20B Chat Organization not supplied | 1,190.78 | 1,183.6 – 1,197.96 | 9,901 |
| 349 | DeepSeek LLM 67B Chat deepseek | 1,185 | 1,173.55 – 1,196.46 | 4,932 |
| 351 | WizardLM 70B microsoft | 1,184.81 | 1,175.4 – 1,194.22 | 8,214 |
| 353 | Yi 34B Chat Organization not supplied | 1,183.78 | 1,176.97 – 1,190.6 | 15,483 |
| 353 | Granite 3.0 8B Instruct ibm | 1,183.21 | 1,174.54 – 1,191.88 | 6,638 |
| 353 | OpenChat 3.5 Organization not supplied | 1,183.11 | 1,173.48 – 1,192.75 | 7,968 |
| 353 | OpenChat 3.5 (0106) Organization not supplied | 1,182.81 | 1,174.83 – 1,190.79 | 12,637 |
| 354 | Gemma 1.1 7B IT google | 1,182.6 | 1,176.54 – 1,188.66 | 23,893 |
| 354 | Snowflake Arctic Instruct Organization not supplied | 1,180.12 | 1,174.22 – 1,186.02 | 32,832 |
| 353 | Granite 3.1 2B Instruct ibm | 1,178.92 | 1,167.75 – 1,190.1 | 3,188 |
| 354 | Tulu 2 DPO 70B Organization not supplied | 1,177.73 | 1,167.94 – 1,187.51 | 6,535 |
| 354 | OpenHermes 2.5 Mistral 7B Organization not supplied | 1,176.03 | 1,165.64 – 1,186.41 | 5,006 |
| 356 | Vicuna 33B Organization not supplied | 1,172.99 | 1,166.81 – 1,179.18 | 22,479 |
| 356 | Phi-3 Small 8K Instruct microsoft | 1,171.13 | 1,165.23 – 1,177.04 | 17,766 |
| 356 | Starling LM 7B Beta Organization not supplied | 1,171.03 | 1,163.64 – 1,178.42 | 16,056 |
| 358 | Llama 2 70B Chat meta | 1,170.7 | 1,165.22 – 1,176.18 | 38,492 |
| 358 | Starling LM 7B Alpha Organization not supplied | 1,167.47 | 1,159.43 – 1,175.52 | 10,224 |
| 361 | Llama 3.2 3B Instruct meta | 1,166.82 | 1,159.12 – 1,174.52 | 7,936 |
| 358 | Nous Hermes 2 Mixtral 8x7B DPO Organization not supplied | 1,164.46 | 1,152.57 – 1,176.34 | 3,777 |
| 370 | Granite 3.0 2B Instruct ibm | 1,156.72 | 1,148.34 – 1,165.1 | 6,837 |
| 366 | Llama 2 70B SteerLM Chat nvidia | 1,154.76 | 1,142.19 – 1,167.33 | 3,585 |
| 367 | QwQ-32B-Preview alibaba | 1,154.25 | 1,142.78 – 1,165.72 | 3,231 |
| 368 | SOLAR 10.7B Instruct v1.0 upstage | 1,152.49 | 1,139.36 – 1,165.62 | 4,155 |
| 366 | Dolphin 2.2.1 Mistral 7B Organization not supplied | 1,152.34 | 1,137.07 – 1,167.61 | 1,679 |
| 371 | MPT 30B Chat Organization not supplied | 1,150.82 | 1,138.62 – 1,163.02 | 2,572 |
| 373 | WizardLM 13B microsoft | 1,149.59 | 1,140.4 – 1,158.78 | 7,044 |
| 373 | Mistral 7B Instruct v0.2 mistral | 1,149.45 | 1,142.82 – 1,156.09 | 19,402 |
| 370 | Falcon 180B Chat Organization not supplied | 1,148.19 | 1,131.19 – 1,165.19 | 1,295 |
| 373 | Qwen1.5-7B-Chat alibaba | 1,143.9 | 1,134.03 – 1,153.77 | 4,737 |
| 374 | Phi-3 Mini 4K Instruct (June 2024) microsoft | 1,143.18 | 1,136.74 – 1,149.61 | 12,297 |
| 375 | Vicuna 13B Organization not supplied | 1,141.63 | 1,134.96 – 1,148.31 | 19,367 |
| 375 | Llama 2 13B Chat meta | 1,141.48 | 1,134.77 – 1,148.19 | 19,174 |
| 374 | Qwen-14B-Chat alibaba | 1,139.42 | 1,128.51 – 1,150.34 | 4,964 |
| 374 | PaLM 2 google | 1,139.24 | 1,129.91 – 1,148.57 | 8,554 |
| 375 | Gemma 7B IT google | 1,137.75 | 1,128.29 – 1,147.22 | 8,925 |
| 375 | Code Llama 34B Instruct meta | 1,136.93 | 1,128.06 – 1,145.79 | 7,366 |
| 379 | Zephyr 7B Beta Organization not supplied | 1,130.87 | 1,122.07 – 1,139.67 | 11,118 |
| 381 | Phi-3 Mini 128K Instruct microsoft | 1,129.9 | 1,122.49 – 1,137.31 | 20,685 |
| 385 | Phi-3 Mini 4K Instruct microsoft | 1,128.16 | 1,121.77 – 1,134.55 | 20,118 |
| 379 | Guanaco 33B Organization not supplied | 1,127.56 | 1,115.51 – 1,139.6 | 2,921 |
| 377 | Zephyr 7B Alpha Organization not supplied | 1,127 | 1,111.46 – 1,142.54 | 1,785 |
| 386 | StripedHyena Nous 7B Organization not supplied | 1,121.68 | 1,110.73 – 1,132.63 | 5,182 |
| 381 | Code Llama 70B Instruct meta | 1,119.22 | 1,101.14 – 1,137.29 | 1,143 |
| 391 | Gemma 1.1 2B IT google | 1,116.7 | 1,108.99 – 1,124.42 | 10,854 |
| 391 | Vicuna 7B Organization not supplied | 1,115.38 | 1,106.28 – 1,124.47 | 6,923 |
| 388 | SmolLM2 1.7B Instruct Organization not supplied | 1,114.76 | 1,100.43 – 1,129.09 | 2,199 |
| 394 | Llama 3.2 1B Instruct meta | 1,111.15 | 1,103.27 – 1,119.03 | 8,045 |
| 394 | Mistral 7B Instruct mistral | 1,110.43 | 1,101.17 – 1,119.68 | 8,977 |
| 395 | Llama 2 7B Chat meta | 1,107.99 | 1,100.95 – 1,115.02 | 14,148 |
| 399 | Gemma 2B IT google | 1,093.69 | 1,082.23 – 1,105.14 | 4,780 |
| 403 | Qwen1.5-4B-Chat alibaba | 1,091.2 | 1,081.89 – 1,100.51 | 7,597 |
| 404 | OLMo 7B Instruct allenai | 1,073.67 | 1,062.55 – 1,084.8 | 6,328 |
| 406 | Koala 13B Organization not supplied | 1,070.99 | 1,061.18 – 1,080.8 | 6,965 |
| 406 | Alpaca 13B Organization not supplied | 1,070.29 | 1,058.91 – 1,081.67 | 5,745 |
| 404 | GPT4All 13B Snoozy Organization not supplied | 1,067.77 | 1,052.64 – 1,082.9 | 1,743 |
| 406 | MPT 7B Chat Organization not supplied | 1,063.51 | 1,051.61 – 1,075.41 | 3,924 |
| 406 | ChatGLM3 6B Organization not supplied | 1,056.63 | 1,045 – 1,068.26 | 4,658 |
| 409 | RWKV-4 Raven 14B Organization not supplied | 1,042.4 | 1,031.09 – 1,053.71 | 4,845 |
| 412 | ChatGLM2 6B Organization not supplied | 1,024.83 | 1,011.3 – 1,038.36 | 2,658 |
| 412 | OpenAssistant Pythia 12B Organization not supplied | 1,023.59 | 1,012.8 – 1,034.38 | 6,310 |
| 415 | ChatGLM 6B Organization not supplied | 995.88 | 983.4 – 1,008.37 | 4,914 |
| 415 | FastChat-T5 3B Organization not supplied | 992.63 | 980.34 – 1,004.92 | 4,203 |
| 415 | Dolly 2.0 12B Organization not supplied | 982.55 | 969.13 – 995.97 | 3,412 |
| 415 | LLaMA 13B meta | 975.24 | 959.58 – 990.9 | 2,391 |
| 418 | StableLM Tuned Alpha 7B Organization not supplied | 953.27 | 940.65 – 965.88 | 3,287 |

## What this ranking can tell you

 - The population is the models in this source board, including variants and models without an available API price.
- Ranks use the published confidence bounds: one plus the number of models whose lower bound exceeds this model’s upper bound. Shared ranks are not proof of equal performance.

## Check the source

The original evidence is retained byte for byte. The downloadable board includes the source URLs, revision and SHA-256 hashes.

[Download this board and provenance →](/data/rankings/arena-text-style-control.json)

 Source files and licences - [metadata · a7a159f557a1](/data/rankings/evidence/a7a159f557a1e594a5d41218fe8d53c63c63daa39b655e51d6fc5b482a17d059.json) · [upstream source ↗](https://huggingface.co/api/datasets/lmarena-ai/leaderboard-dataset)
- [licence · 88a46d472f35](/data/rankings/evidence/88a46d472f35274fb6b7a852c3d16c8ee8007d51b37f290161537bec26895ee4.txt) · [upstream source ↗](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/raw/8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53/README.md)
- [rows · 7e575fc4ab7c](/data/rankings/evidence/7e575fc4ab7c78f75d02c4ec86dba4d5ecbfc02f21e14372fe22de1113d952ee.parquet) · [upstream source ↗](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/resolve/8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53/text_style_control/latest-00000-of-00001.parquet)

[Explore other model types →](/rankings/)[Read the benchmark guide →](/benchmarks/)

## Continue your investigation

 - [Understand the measurements](/benchmarks/)
- [Compare task performance](/best/)
- [Compare embedding models](/embeddings/)
- [Inspect ranking rules](/methodology/)