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

# Agents · explicit task outcomes: model rankings and source evidence · Undominated.ai

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

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

# Agents · explicit task outcomes

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 · explicit task outcomes · IPS score

| Significance rank | Model / evaluated variant | IPS score | 95% confidence interval | Observations |
| --- | --- | --- | --- | --- |
| 1 | Claude Opus 5.5 (Max) anthropic | 0.211494 | 0.179358 – 0.243629 | 4,352 |
| 1 | Claude Sonnet 5.5 (Max) anthropic | 0.202096 | 0.158045 – 0.246146 | 2,205 |
| 1 | Gemini 4 Carbon (High) google | 0.180213 | 0.16765 – 0.192776 | 22,690 |
| 2 | GPT-6 Astra (Max) openai | 0.141685 | 0.108107 – 0.175263 | 6,555 |
| 4 | Gemini 4 Argon (High) google | 0.133671 | 0.118328 – 0.149013 | 26,963 |
| 2 | Claude Fable 5.1 (Max) anthropic | 0.132905 | 0.097311 – 0.168498 | 7,487 |
| 4 | GPT-6.1 Sol (Max) openai | 0.121895 | 0.102325 – 0.141466 | 12,235 |
| 4 | Muse Wonder Full (Max) meta | 0.114416 | 0.096455 – 0.132377 | 16,254 |
| 4 | Claude Opus 5 (Max) anthropic | 0.113834 | 0.090077 – 0.13759 | 18,539 |
| 4 | GPT-6.1 Terra (Max) openai | 0.111336 | 0.083964 – 0.138707 | 10,041 |
| 6 | Hy4 tencent | 0.073951 | 0.041718 – 0.106184 | 4,692 |
| 9 | Muse Wonder (Max) meta | 0.063307 | 0.034035 – 0.092579 | 8,073 |
| 11 | Claude Fable 5 (High) anthropic | 0.056483 | 0.034532 – 0.078434 | 36,809 |
| 11 | MiMo-V2.6-Flash xiaomi | 0.055039 | 0.032476 – 0.077601 | 13,651 |
| 10 | Grok 4.7 (xHigh) xai | 0.053406 | 0.02237 – 0.084442 | 9,073 |
| 11 | MiMo-V2.6-Pro xiaomi | 0.052491 | 0.033002 – 0.07198 | 21,044 |
| 11 | Hy4 preview tencent | 0.045471 | 0.029889 – 0.061054 | 36,269 |
| 11 | Qwen3.8-Max alibaba | 0.043377 | 0.029049 – 0.057704 | 42,576 |
| 11 | Kimi K3.1 Pro (Max) moonshot | 0.040177 | 0.011919 – 0.068436 | 10,202 |
| 11 | Muse Spark 1.3 (Max) meta | 0.039782 | 0.021949 – 0.057615 | 44,614 |
| 11 | Kimi K3 (Max) moonshot | 0.037458 | 0.024496 – 0.050421 | 99,584 |
| 11 | GPT-6 Sol (Max) openai | 0.035376 | 0.001195 – 0.069556 | 6,126 |
| 11 | DeepSeek-V4.1-Flash-20261017 (Max) deepseek | 0.033771 | 0.020559 – 0.046982 | 35,025 |
| 11 | GLM-5.3 (Max) zai | 0.031922 | 0.014407 – 0.049436 | 58,030 |
| 11 | GLM-5.2 (Max) zai | 0.025385 | 0.00849 – 0.042279 | 57,302 |
| 12 | GLM-5.3-Flash zai | 0.02316 | 0.007056 – 0.039264 | 59,023 |
| 11 | GPT-5.6 Sol (xHigh) openai | 0.019726 | -0.002776 – 0.042229 | 31,493 |
| 12 | Claude Haiku 5.5 (High) anthropic | 0.014338 | -0.010339 – 0.039014 | 16,264 |
| 17 | Gemini 3.8 Flash (High) google | 0.011985 | -0.005741 – 0.029711 | 28,414 |
| 11 | MiniMax-M3.1-Flash minimax | 0.010977 | -0.026916 – 0.048871 | 4,626 |
| 21 | Qwen3.8-Flash-Next alibaba | 0.003862 | -0.013584 – 0.021307 | 41,035 |
| 18 | Claude Sonnet 5 (High) anthropic | -0.00023 | -0.02637 – 0.02591 | 27,267 |
| 25 | GPT-6 Luna (Max) openai | -0.001673 | -0.011665 – 0.008318 | 110,526 |
| 18 | Step 5 Preview stepfun | -0.004597 | -0.036741 – 0.027546 | 7,033 |
| 25 | Grok 4.6 (xHigh) xai | -0.018889 | -0.045235 – 0.007456 | 17,841 |
| 28 | Qwen3.8-27B alibaba | -0.02419 | -0.043734 – -0.004646 | 31,813 |
| 24 | DeepSeek-V4-Pro (High, 20260813) deepseek | -0.027969 | -0.066191 – 0.010253 | 3,402 |
| 31 | Gemini 3.7 Flash (High) google | -0.031184 | -0.050586 – -0.011782 | 43,575 |
| 26 | Mistral Large 4 mistral | -0.04311 | -0.089689 – 0.00347 | 2,389 |
| 37 | GPT-6 Luna (Medium) openai | -0.062334 | -0.075377 – -0.049292 | 36,762 |
| 37 | Muse Spark 1.1 meta | -0.065335 | -0.080794 – -0.049876 | 86,947 |
| 38 | GPT-5.6 Terra (xHigh) openai | -0.073179 | -0.091222 – -0.055135 | 29,566 |
| 35 | Spark-X2.6 synlan | -0.078792 | -0.116422 – -0.041162 | 4,124 |
| 39 | Muse Spark 1.2 (xHigh) meta | -0.091773 | -0.113193 – -0.070353 | 30,983 |
| 38 | Qwen3.7-Plus alibaba | -0.092098 | -0.128426 – -0.05577 | 15,743 |
| 40 | Qwen3.7-Max alibaba | -0.103331 | -0.130636 – -0.076027 | 28,968 |
| 37 | Beam 1.0 501B A23B reflection | -0.106308 | -0.1649 – -0.047717 | 5,050 |
| 41 | MiMo-V2.5-Pro xiaomi | -0.108486 | -0.132301 – -0.084671 | 29,180 |
| 41 | Hy3 tencent | -0.116695 | -0.147749 – -0.085642 | 13,853 |
| 42 | Gemini 3.1 Pro Preview google | -0.119168 | -0.147388 – -0.090947 | 69,499 |
| 48 | GPT-6 Luna (Low) openai | -0.152584 | -0.170912 – -0.134256 | 22,963 |
| 46 | Mistral Medium 3.5 mistral | -0.170482 | -0.210413 – -0.130552 | 7,353 |
| 50 | MiniMax-M3 minimax | -0.176701 | -0.20539 – -0.148011 | 26,992 |
| 50 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B (NVFP4) nvidia | -0.186147 | -0.209256 – -0.163037 | 10,821 |
| 50 | Nemotron 3 Ultra nvidia | -0.233662 | -0.310611 – -0.156712 | 4,275 |
| 55 | Inkling-Small thinky | -0.264232 | -0.31302 – -0.215444 | 7,078 |
| 55 | Inkling thinky | -0.270384 | -0.309389 – -0.231379 | 24,084 |
| 55 | Solar Pro 4 upstage | -0.307814 | -0.366177 – -0.249452 | 4,588 |

## 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-task-outcome-explicit.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 · 5e92ac469b52](/data/rankings/evidence/5e92ac469b5228e8dd69af6e80b180527dc30728d814c5a8ec7cf17f7b8e1bd1.parquet) · [upstream source ↗](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/resolve/8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53/agent_task_outcome_explicit/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/)