RankingLMArena Agent Arena

LMArena Agent Arena

Data updated 22 Nov 2026

LMArena Agent Arena is published here as its own result table, version 2026-10-02. Scores use index, and higher values rank better. The display harness is LMArena Agent Arena · max. Data updated 22 Nov 2026.

Bucket
Supporting evidence
Unit
index
Direction
Higher is better
Version
2026-10-02
Display harness
LMArena Agent Arena · max
Board
https://arena.ai/leaderboard/agent

These rows are the aggregate Agent Arena IPS net improvement, in percentage points, with the board's session count. The text arena's Bradley-Terry rating is a different column. These rows stay out of the weighted Overall.

Compare published benchmark results with category weights →

Models

Results are ordered by score in the display harness, followed by models with no result. Other harnesses are listed separately below. Missing results remain unknown.

426–450 of 593 entries

ModelScoreEvidenceSource-recorded date
NVIDIA-Nemotron-Nano-12B-v2-VL-BF16NVIDIA———
NVIDIA-Nemotron-Nano-9B-v2NVIDIA———
NVIDIA-Nemotron-Nano-9B-v2-JapaneseNVIDIA———
o1OpenAI———
O1 MiniOpenAI———
O1 PreviewOpenAI———
o1-proOpenAI———
o3OpenAI———
o3-miniOpenAI———
o3-proOpenAI———
o4-miniOpenAI———
Phi-3 Mini Instruct 3.8BMicrosoft———
Phi-3-medium-128k-instructMicrosoft———
Phi-3-medium-4k-instructMicrosoft———
Phi-3-mini-128k-instructMicrosoft———
Phi-3-mini-4k-instructMicrosoft———
Phi-3-small-128k-instructMicrosoft———
Phi-3-small-8k-instructMicrosoft———
Phi-3-vision-128k-instructMicrosoft———
Phi-3.5-mini-instructMicrosoft———
Phi-3.5-MoE-instructMicrosoft———
Phi-3.5-vision-instructMicrosoft———
phi-4Microsoft———
Phi-4-mini-flash-reasoningMicrosoft———
Phi-4-mini-instructMicrosoft———

Other harnesses

These runs use other harnesses. Capability scoring matches configurations separately; model details identify which source records contribute.

ModelScoreHarnessEvidenceSource-recorded date
Claude Fable 5Anthropic5.86 LMArena Agent Arena · highofficial board2026-11-18
Claude Haiku 5.5Anthropic-0.61 LMArena Agent Arena · highofficial board2026-11-18
Claude Opus 4.8Anthropic7.14 LMArena Agent Arena · highofficial board2026-10-14
Claude Opus 5Anthropic7.95 LMArena Agent Arena · highofficial board2026-10-08
Claude Opus 5.5Anthropic13.7 LMArena Agent Arena · highofficial board2026-11-18
Claude Sonnet 5Anthropic1 LMArena Agent Arena · highofficial board2026-11-18
Claude Sonnet 5.5Anthropic7.42 LMArena Agent Arena · highofficial board2026-11-18
DeepSeek V4-Pro 0813DeepSeek0.36 LMArena Agent Arena · highofficial board2026-10-14
Gemini 3.1 ProGoogle-5.69 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Gemini 3.6 FlashGoogle-4.72 LMArena Agent Arena · highofficial board2026-10-14
Gemini 3.7 FlashGoogle-3.07 LMArena Agent Arena · highofficial board2026-11-18
Gemini 3.8 FlashGoogle1.49 LMArena Agent Arena · highofficial board2026-11-18
Gemini 4 ArgonGoogle10.31 LMArena Agent Arena · highofficial board2026-11-18
GLM-5.3-FlashZ.ai-3.87 LMArena Agent Arena · unspecifiedofficial board2026-11-18
GPT-5.6 SolOpenAI3.68 LMArena Agent Arena · xhighofficial board2026-11-18
GPT-5.6 TerraOpenAI-2.67 LMArena Agent Arena · xhighofficial board2026-11-18
Grok 4.6xAI-0.6 LMArena Agent Arena · xhighofficial board2026-11-18
Grok 4.7xAI1.48 LMArena Agent Arena · xhighofficial board2026-11-18
Grok-4.5xAI0.07 LMArena Agent Arena · unspecifiedofficial board2026-10-14
Hy3Tencent-8.63 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Hy4-previewTencent-0.44 LMArena Agent Arena · unspecifiedofficial board2026-11-18
MiMo-V2.6-FlashXiaomi-1.75 LMArena Agent Arena · unspecifiedofficial board2026-11-18
MiMo-V2.6-ProXiaomi1.42 LMArena Agent Arena · unspecifiedofficial board2026-11-18
MiniMax-M3MiniMax-10.92 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Mistral Large 4Mistral-7.15 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Mistral Medium 3.5Mistral-14.46 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Muse Spark 1.1Meta-9.05 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Muse Spark 1.2Meta-9.11 LMArena Agent Arena · xhighofficial board2026-11-18
NVIDIA Nemotron 3 UltraNVIDIA-16.8 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Qwen 3.8-MaxQwen-0.91 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Qwen3.7-maxQwen-8.25 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Qwen3.7-plusQwen-8.98 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Qwen3.8-27BQwen-5.5 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Qwen3.8-Flash-NextQwen-4.91 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Solar Pro 4Upstage-23.89 LMArena Agent Arena · unspecifiedofficial board2026-11-18
Step 5 PreviewStepFun-5.15 LMArena Agent Arena · unspecifiedofficial board2026-11-18