---
title: "Agents · praise and complaints: model rankings and source evidence · Undominated.ai"
canonical: https://undominated.ai/rankings/arena-agent-praise-complaint/
description: "Agents · praise and complaints ranked by IPS score. Inspect exact model variants, source dates, measurement limits and downloadable evidence."
---

# Agents · praise and complaints: model rankings and source evidence · Undominated.ai

> Agents · praise and complaints ranked by IPS score. Inspect exact model variants, source dates, measurement limits and downloadable evidence.

[All rankings](/rankings/) / Coding and agents

# Agents · praise and complaints

Agent Arena signal scores from the official dataset. Each signal is a separate measurement; it is not a success percentage.

**58 entries** · IPS score · Higher is better

Source publication: 2026-11-04. 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.

Agents · praise and complaints · IPS score

| Significance rank | Model / evaluated variant | IPS score | 95% confidence interval | Observations |
| --- | --- | --- | --- | --- |
| 1 | Claude Opus 5.5 (Max) anthropic | 0.429138 | 0.350451 – 0.507825 | 1,761 |
| 1 | Gemini 4 Carbon (High) google | 0.401204 | 0.373168 – 0.42924 | 10,208 |
| 1 | GPT-6 Astra (Max) openai | 0.327426 | 0.251388 – 0.403463 | 2,577 |
| 1 | Claude Sonnet 5.5 (Max) anthropic | 0.297119 | 0.189183 – 0.405054 | 906 |
| 3 | Claude Fable 5.1 (Max) anthropic | 0.279718 | 0.210859 – 0.348576 | 3,286 |
| 3 | GPT-6.1 Sol (Max) openai | 0.27418 | 0.23419 – 0.31417 | 5,246 |
| 3 | Gemini 4 Argon (High) google | 0.228661 | 0.199718 – 0.257603 | 12,485 |
| 5 | GPT-6.1 Terra (Max) openai | 0.17316 | 0.124935 – 0.221385 | 5,399 |
| 7 | Muse Wonder Full (Max) meta | 0.162615 | 0.132263 – 0.192967 | 8,487 |
| 8 | Claude Fable 5 (High) anthropic | 0.150519 | 0.114535 – 0.186503 | 14,520 |
| 8 | Claude Opus 5 (Max) anthropic | 0.119239 | 0.078104 – 0.160375 | 8,202 |
| 8 | GPT-6 Sol (Max) openai | 0.104567 | 0.053437 – 0.155698 | 2,974 |
| 10 | Kimi K3.1 Pro (Max) moonshot | 0.084586 | 0.044398 – 0.124773 | 5,688 |
| 8 | Hy4 tencent | 0.073277 | 0.012746 – 0.133807 | 2,084 |
| 11 | GPT-5.6 Sol (xHigh) openai | 0.057136 | 0.024533 – 0.08974 | 13,030 |
| 12 | Gemini 3.8 Flash (High) google | 0.052642 | 0.027479 – 0.077806 | 14,974 |
| 12 | MiMo-V2.6-Pro xiaomi | 0.047531 | 0.019632 – 0.075429 | 11,318 |
| 12 | GLM-5.3 (Max) zai | 0.043749 | 0.019977 – 0.067521 | 30,739 |
| 12 | Grok 4.7 (xHigh) xai | 0.028642 | -0.016926 – 0.074209 | 4,637 |
| 13 | GLM-5.2 (Max) zai | 0.02317 | 0.000209 – 0.046131 | 25,273 |
| 13 | Muse Spark 1.3 (Max) meta | 0.021459 | -0.003868 – 0.046787 | 22,469 |
| 14 | DeepSeek-V4.1-Flash-20261017 (Max) deepseek | 0.019134 | 0.001722 – 0.036546 | 20,235 |
| 14 | Kimi K3 (Max) moonshot | 0.013673 | -0.00356 – 0.030905 | 47,105 |
| 12 | DeepSeek-V4-Pro (High, 20260813) deepseek | 0.007327 | -0.044369 – 0.059022 | 1,608 |
| 16 | Qwen3.8-Max alibaba | 0.002303 | -0.017554 – 0.022159 | 21,636 |
| 16 | Hy4 preview tencent | 0.002149 | -0.019223 – 0.02352 | 18,086 |
| 14 | Muse Wonder (Max) meta | -0.002547 | -0.043264 – 0.03817 | 4,607 |
| 15 | MiMo-V2.6-Flash xiaomi | -0.004816 | -0.036304 – 0.026671 | 6,895 |
| 16 | Claude Sonnet 5 (High) anthropic | -0.010621 | -0.04571 – 0.024468 | 10,020 |
| 20 | GPT-6 Luna (Max) openai | -0.011909 | -0.025529 – 0.00171 | 51,440 |
| 19 | GPT-5.6 Terra (xHigh) openai | -0.015536 | -0.03881 – 0.007739 | 13,527 |
| 16 | Gemini 3.1 Pro Preview google | -0.01573 | -0.052227 – 0.020768 | 27,127 |
| 18 | Grok 4.6 (xHigh) xai | -0.018784 | -0.052966 – 0.015398 | 9,204 |
| 19 | Gemini 3.7 Flash (High) google | -0.021198 | -0.045039 – 0.002643 | 23,463 |
| 26 | GLM-5.3-Flash zai | -0.041176 | -0.060267 – -0.022084 | 34,330 |
| 23 | Claude Haiku 5.5 (High) anthropic | -0.041512 | -0.07253 – -0.010493 | 9,012 |
| 23 | Step 5 Preview stepfun | -0.05228 | -0.09329 – -0.01127 | 3,676 |
| 27 | Qwen3.8-Flash-Next alibaba | -0.055707 | -0.076563 – -0.034852 | 20,972 |
| 18 | Beam 1.0 501B A23B reflection | -0.058608 | -0.130476 – 0.013259 | 2,370 |
| 27 | Qwen3.8-27B alibaba | -0.059754 | -0.084265 – -0.035243 | 16,954 |
| 34 | GPT-6 Luna (Medium) openai | -0.067973 | -0.08351 – -0.052436 | 19,400 |
| 27 | MiniMax-M3.1-Flash minimax | -0.075768 | -0.121125 – -0.030411 | 2,768 |
| 33 | Hy3 tencent | -0.087397 | -0.124632 – -0.050163 | 5,277 |
| 36 | Qwen3.7-Max alibaba | -0.094479 | -0.12537 – -0.063587 | 13,141 |
| 40 | GPT-6 Luna (Low) openai | -0.104209 | -0.123519 – -0.0849 | 12,382 |
| 37 | Qwen3.7-Plus alibaba | -0.113007 | -0.151859 – -0.074156 | 6,885 |
| 36 | Mistral Large 4 mistral | -0.118175 | -0.170074 – -0.066276 | 1,228 |
| 41 | Muse Spark 1.2 (xHigh) meta | -0.128443 | -0.149827 – -0.10706 | 17,621 |
| 41 | Muse Spark 1.1 meta | -0.134359 | -0.149585 – -0.119133 | 43,694 |
| 41 | MiMo-V2.5-Pro xiaomi | -0.140347 | -0.164706 – -0.115989 | 13,202 |
| 46 | MiniMax-M3 minimax | -0.160778 | -0.187038 – -0.134518 | 12,413 |
| 46 | Mistral Medium 3.5 mistral | -0.185879 | -0.222649 – -0.149108 | 3,307 |
| 41 | Nemotron 3 Ultra nvidia | -0.190939 | -0.263249 – -0.118629 | 1,290 |
| 49 | Spark-X2.6 synlan | -0.194626 | -0.226611 – -0.162641 | 2,513 |
| 52 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4) nvidia | -0.222581 | -0.241094 – -0.204067 | 6,427 |
| 52 | Inkling-Small thinky | -0.227703 | -0.263055 – -0.192351 | 3,487 |
| 52 | Solar Pro 4 upstage | -0.24175 | -0.281444 – -0.202056 | 2,314 |
| 52 | Inkling thinky | -0.246344 | -0.272172 – -0.220516 | 10,406 |

## 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-agent-praise-complaint.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 · 22c80107963a](/data/rankings/evidence/22c80107963a734288b84edaf6045ae59a6ece9ab73a44d83633d906fd9396c4.parquet) · [upstream source ↗](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/resolve/8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53/agent_praise_complaint/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/)