Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
LMArena · Agent · Task Outcome
LMArena's Task Outcome board: models ranked by people voting between two anonymous agent sessions. Every model LMArena ranks is listed; those in the ModelChoice catalog link to their page.
Each point is a model with a published LMArena Task Outcome value and complete token prices, on the 75% input / 25% output blend. Frontier models are spaced evenly along the cost axis.
68 models, as LMArena published them on 2026-11-18.
| Rank | Model | Maker | Score | 95% interval | Sessions | Licence |
|---|---|---|---|---|---|---|
| 1 | Anthropic | 0.225 | 0.191–0.258 | 3,495 | Proprietary | |
| 2 | Anthropic | 0.192 | 0.159–0.226 | 3,830 | Proprietary | |
| 3 | 0.192 | 0.155–0.229 | 2,134 | Proprietary | ||
| 4 | Anthropic | 0.176 | 0.129–0.223 | 2,004 | Proprietary | |
| 5 | 0.172 | 0.160–0.183 | 23,115 | Proprietary | ||
| 6 | 0.157 | 0.128–0.186 | 3,482 | Proprietary | ||
| 7 | OpenAI | 0.149 | 0.114–0.183 | 5,641 | Proprietary | |
| 8 | Anthropic | 0.142 | 0.105–0.179 | 5,451 | Proprietary | |
| 9 | 0.126 | 0.109–0.143 | 20,165 | Proprietary | ||
| 10 | Anthropic | 0.126 | 0.106–0.146 | 9,362 | Proprietary | |
| 11 | OpenAI | 0.124 | 0.081–0.168 | 1,584 | Proprietary | |
| 12 | OpenAI | 0.116 | 0.095–0.136 | 11,070 | Proprietary | |
| 13 | OpenAI | 0.109 | 0.069–0.150 | 1,935 | Proprietary | |
| 14 | Meta | 0.104 | 0.086–0.122 | 12,152 | Proprietary | |
| 15 | Anthropic | 0.101 | 0.075–0.127 | 10,806 | Proprietary | |
| 16 | Alibaba | 0.088 | 0.063–0.114 | 4,928 | Proprietary | |
| 17 | OpenAI | 0.085 | 0.062–0.109 | 10,685 | Proprietary | |
| 18 | Anthropic | 0.071 | 0.031–0.110 | 4,251 | Proprietary | |
| 19 | Tencent | 0.063 | 0.039–0.088 | 8,870 | Proprietary | |
| 20 | OpenAI | 0.056 | 0.035–0.077 | 9,828 | Proprietary | |
| 21 | Anthropic | 0.044 | 0.020–0.069 | 20,358 | Proprietary | |
| 22 | OpenAI | 0.039 | 0.005–0.073 | 5,001 | Proprietary | |
| 23 | Alibaba | 0.036 | 0.020–0.052 | 29,811 | Proprietary | |
| 24 | xAI | 0.035 | 0.002–0.069 | 6,380 | Proprietary | |
| 25 | Xiaomi | 0.035 | 0.015–0.055 | 18,420 | MIT | |
| 26 | Tencent | 0.032 | 0.016–0.049 | 29,061 | Apache 2.0 | |
| 27 | Moonshot AI | 0.027 | 0.013–0.042 | 58,199 | Kimi K3 license | |
| 28 | Z.AI | 0.027 | 0.008–0.046 | 33,859 | MIT | |
| 29 | OpenAI | 0.025 | 0.002–0.049 | 18,123 | Proprietary | |
| 30 | Meta | 0.025 | 0.005–0.045 | 32,348 | Proprietary | |
| 31 | Xiaomi | 0.025 | 0.004–0.046 | 12,962 | MIT | |
| 32 | Meta | 0.025 | -0.005–0.055 | 6,311 | Proprietary | |
| 33 | TMInkling 2 Small | Thinking Machines Lab | 0.024 | 0.002–0.047 | 7,716 | Apache 2.0 |
| 34 | DeepSeek | 0.022 | 0.011–0.034 | 35,972 | MIT | |
| 35 | Moonshot AI | 0.019 | -0.005–0.044 | 8,948 | Kimi K3.1 Pro License | |
| 36 | Z.AI | 0.012 | -0.007–0.030 | 33,548 | MIT | |
| 37 | 0.010 | -0.009–0.029 | 19,321 | Proprietary | ||
| 38 | Z.AI | 0.006 | -0.011–0.024 | 37,787 | MIT | |
| 39 | OpenAI | 0.004 | -0.009–0.017 | 23,173 | Proprietary | |
| 40 | Alibaba | 0.003 | -0.016–0.021 | 32,469 | qwen-community-1.0 | |
| 41 | Anthropic | 0.002 | -0.023–0.027 | 12,371 | Proprietary | |
| 42 | Anthropic | -0.009 | -0.038–0.019 | 15,937 | Proprietary | |
| 43 | OpenAI | -0.012 | -0.023–-0.001 | 89,825 | Proprietary | |
| 44 | Meta | -0.020 | -0.040–0.000 | 10,288 | Proprietary | |
| 45 | StepFun | -0.021 | -0.055–0.012 | 5,790 | Proprietary | |
| 46 | xAI | -0.025 | -0.054–0.005 | 11,290 | Proprietary | |
| 47 | MiniMax | -0.026 | -0.062–0.010 | 4,590 | MiniMax Community License | |
| 48 | Alibaba | -0.033 | -0.054–-0.011 | 23,183 | Apache 2.0 | |
| 49 | -0.034 | -0.055–-0.014 | 25,842 | Proprietary | ||
| 50 | Meta | -0.073 | -0.088–-0.057 | 56,355 | Proprietary | |
| 51 | Mistral | -0.075 | -0.127–-0.023 | 2,193 | Modified MIT | |
| 52 | OpenAI | -0.076 | -0.087–-0.066 | 40,117 | Proprietary | |
| 53 | OpenAI | -0.077 | -0.097–-0.058 | 18,390 | Proprietary | |
| 54 | RABeam 1.0 501B A23B | Reflection AI | -0.089 | -0.152–-0.026 | 4,180 | Apache 2.0 |
| 55 | Meta | -0.101 | -0.125–-0.078 | 21,769 | Proprietary | |
| 56 | Alibaba | -0.106 | -0.136–-0.075 | 16,429 | Proprietary | |
| 57 | -0.117 | -0.149–-0.085 | 34,466 | Proprietary | ||
| 58 | SYSpark-X2.6 | synlan | -0.119 | -0.149–-0.089 | 5,660 | Proprietary |
| 59 | Alibaba | -0.119 | -0.159–-0.080 | 9,943 | Proprietary | |
| 60 | Tencent | -0.123 | -0.157–-0.090 | 10,509 | Apache 2.0 | |
| 61 | OpenAI | -0.146 | -0.165–-0.128 | 20,400 | Proprietary | |
| 62 | MiniMax | -0.175 | -0.205–-0.144 | 17,005 | MiniMax Community License | |
| 63 | Mistral | -0.175 | -0.216–-0.134 | 5,474 | Modified MIT | |
| 64 | NVIDIA | -0.191 | -0.209–-0.172 | 15,416 | OpenMDW-1.1 | |
| 65 | NVIDIA | -0.237 | -0.315–-0.160 | 3,274 | OpenMDW-1.1 | |
| 66 | TMInkling | Thinking Machines Lab | -0.269 | -0.310–-0.228 | 16,156 | Apache 2.0 |
| 67 | TMInkling-Small | Thinking Machines Lab | -0.272 | -0.322–-0.222 | 5,391 | Apache 2.0 |
| 68 | Upstage | -0.318 | -0.379–-0.257 | 3,841 | Proprietary |
Source: LMArena, CC BY 4.0. Read once a day. All leaderboards