AgMoDB
ModelsAgentsOrchestratorsEvalsCompositesVisualizeIndustry
AgMoDB by @mistakeknot
All Benchmarks

Chatbot Arena ELO

Chatbot Arena ELO measures relative user preference from blind head-to-head votes in LMArena. It is a live human-judgment signal for conversational quality under real prompts.

aggregatechatbot_arena

Strengths

Large-scale blinded pairwise voting reduces single-rater bias; continuous updates surface quality shifts quickly after model releases.

Caveats

Voter population is self-selected and may not match enterprise or domain users; style, verbosity, and safety tone can influence votes independently of factual correctness. Data: Arena (lmarena.ai) leaderboard dataset, https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset, licensed CC-BY-4.0; rows are converted to the ELO fields shown here.

How to interpret scores

Contamination risk: lowFreshness: continuous

Higher is better. Scores are ELO-like ratings (typically 900-1400). Compare relative ranking — absolute values shift as new models enter.

Relevant use cases

Model selectionQuality comparisonHuman preference tracking

Caveat to keep in mind

Voter population is self-selected and may not match enterprise or domain users.

Leaderboard

Last synced: Nov 22, 2026, 8:18 AM
#ModelProviderScorePrice/M
1
Gemma 3 12B Instruct
source chatbot_arena
Evidence observed Feb 12, 2026
Google4203.0$0.00
2
Ministral 3 14B
source chatbot_arena
Evidence observed Feb 12, 2026
Mistral AI3412.0$0.20
3
OpenAI: GPT-3.5 Turbo 16k
source chatbot_arena
Evidence observed Feb 12, 2026
OpenAI2391.0$3.25
4
Gemini 4 Carbon (High)
source chatbot_arena
Evidence observed Nov 22, 2026
Google1539.0$8.00
5
Gemini 4 Argon (High)
source chatbot_arena
Evidence observed Nov 22, 2026
Google1528.0$8.00
6
Claude Opus 5.5
source chatbot_arena
Evidence observed Nov 22, 2026
Anthropic1513.0$8.00
7
Claude Fable 5.1 (Max, Default Fallback)
source chatbot_arena
Evidence observed Nov 22, 2026
Anthropic1512.0$20.00
8
Claude Opus 5 (Max)
source chatbot_arena
Evidence observed Nov 22, 2026
Anthropic1506.0$10.00
9
Claude Opus 5.5 (High, Default Fallback)
source chatbot_arena
Evidence observed Oct 8, 2026
Anthropic1504.0$8.00
10
Muse Wonder (Max)
source chatbot_arena
Evidence observed Nov 22, 2026
Meta1503.0$8.00
11
Claude Opus 5 (High)
source chatbot_arena
Evidence observed Nov 22, 2026
Anthropic1502.0$10.00
12
Claude Opus 4.6 (Max)
source chatbot_arena
Evidence observed Mar 26, 2026
Anthropic1502.0$10.00
13
Claude Opus 4.6 (Non-reasoning, High)
source chatbot_arena
Evidence observed Nov 22, 2026
Anthropic1501.0$10.00
14
Gemini 3.8 Flash (High)
source chatbot_arena
Evidence observed Nov 22, 2026
Google1500.0$1.50
15
Qwen: Qwen3.8 Max
source chatbot_arena
Preliminary
Evidence observed Aug 3, 2026
Alibaba1496.0$3.00
16
Muse Spark 1.3 (max)
source chatbot_arena
Evidence observed Nov 22, 2026
Meta1494.0$2.00
17
Muse Spark 1.3 (Max)
source chatbot_arena
Evidence observed Oct 8, 2026
Meta1494.0$2.00
18
MiMo-V2.6-Pro
source chatbot_arena
Evidence observed Nov 22, 2026
Xiaomi1493.0$0.54
19
Claude Fable 5 (Max, Opus 4.8 Fallback)
source chatbot_arena
Evidence observed Nov 22, 2026
Anthropic1492.0$20.00
20
Claude Opus 4.7
source chatbot_arena
Evidence observed Nov 22, 2026
Anthropic1490.0$10.00
21
Gemini 3.7 Flash (High)
source chatbot_arena
Evidence observed Nov 22, 2026
Google1489.0$1.50
22
Qwen3.8 Max Preview
source chatbot_arena
Evidence observed Nov 22, 2026
Alibaba1485.0$3.00
23
GPT-5.4 (Non-reasoning)
source chatbot_arena
Evidence observed Mar 26, 2026
OpenAI1485.0$5.63
24
Muse Spark 1.2 (Xhigh)
source chatbot_arena
Evidence observed Nov 22, 2026
Meta1484.0$2.00
25
Gemini 3.5 Flash (High)
source chatbot_arena
Evidence observed Nov 22, 2026
Google1483.0$3.38
26
GPT-5.2 (Medium)
source chatbot_arena
Evidence observed Mar 26, 2026
OpenAI1482.0$4.81
27
GPT-5.2 (Non-reasoning)
source chatbot_arena
Evidence observed Mar 26, 2026
OpenAI1482.0$4.81
28
Qwen3.8 Max
source chatbot_arena
Evidence observed Sep 15, 2026
Alibaba1481.0$3.00
29
Grok 4.20 0309 (Non-reasoning)
source chatbot_arena
Evidence observed Mar 26, 2026
SpaceXAI1481.0$3.00
30
Gemini 3.6 Flash (High)
source chatbot_arena
Evidence observed Nov 22, 2026
Google1480.0$1.50
31
Gemini 3 Pro Preview (High)
source chatbot_arena
Evidence observed Nov 22, 2026
Google1480.0$4.50
32
Gemini 3.1 Pro Preview
source chatbot_arena
Evidence observed Nov 22, 2026
Google1480.0$4.50
33
Muse Spark 1.1 (Xhigh)
source chatbot_arena
Evidence observed Nov 22, 2026
Meta1479.0$2.00
34
Gemini 3.5 Flash (Medium)
source chatbot_arena
Evidence observed Nov 22, 2026
Google1478.0$3.38
35
Kimi K3
source chatbot_arena
Evidence observed Nov 22, 2026
Kimi1475.0$6.00
36
GLM-5.3 (Max)
source chatbot_arena
Evidence observed Nov 22, 2026
Z AI1475.0$2.15
37
Qwen3.7 Max
source chatbot_arena
Evidence observed Nov 22, 2026
Alibaba1475.0$3.75
38
Muse Spark
source chatbot_arena
Evidence observed Nov 22, 2026
Meta1474.0$0.00
39
GLM-5.3-Flash
source chatbot_arena
Evidence observed Nov 22, 2026
Z AI1471.0$0.24
40
GPT-5.5 (High)
source chatbot_arena
Evidence observed Nov 22, 2026
OpenAI1471.0$11.25
41
Claude Sonnet 5.5 (Xhigh, Default Fallback)
source chatbot_arena
Evidence observed Oct 21, 2026
Anthropic1470.0$4.00
42
Claude Sonnet 5.5
source chatbot_arena
Evidence observed Nov 22, 2026
Anthropic1470.0$4.00
43
GPT-5.4 (Xhigh)
source chatbot_arena
Evidence observed Nov 22, 2026
OpenAI1469.0$5.63
44
GLM-5.2 (Max)
source chatbot_arena
Evidence observed Nov 22, 2026
Z AI1469.0$2.15
45
Gemini 3 Flash Preview (Reasoning)
source chatbot_arena
Evidence observed Nov 22, 2026
Google1466.0$1.13
46
GPT-5.5 (Xhigh)
source chatbot_arena
Evidence observed Nov 22, 2026
OpenAI1466.0$11.25
47
MiMo-V2.5-Pro (Reasoning)
source chatbot_arena
Evidence observed Nov 22, 2026
Xiaomi1464.0$0.54
48
GLM-5.1 (Reasoning)
source chatbot_arena
Evidence observed Nov 22, 2026
Z AI1463.0$1.83
49
DeepSeek V4.1 Flash (Max)
source chatbot_arena
Evidence observed Nov 22, 2026
DeepSeek1463.0$0.52
50
Grok 4.1 Fast (Non-reasoning)
source chatbot_arena
Evidence observed Feb 19, 2026
SpaceXAI1462.0$0.00