Skip to content

Benchmark

Arena ELO - Russian

People compare model answers and choose which they prefer. Higher scores are better.

Models measured
332
Scores recorded
387
Sources
1
IQ weight
1.42

Score distribution

1,322.5
Mean
1,355.7
Median
134.6
Std dev
1,109.2
10th percentile
1,466.9
90th percentile
936.6
Min
936.61,560.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.42 for chat models, measured over 332 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51Grok 4.20Reasoning, 0309, beta· LMArena data1,457.3
52GPT-6.1 SolMax· LMArena data1,454.5
53GLM 5.1· LMArena data1,454.1
54Gemini 3 FlashMinimal· LMArena data1,453.7
55DeepSeek V4 Pro· LMArena data1,453.1
56GPT-5.6 TerraxHigh· LMArena data1,452.9
57Grok 4.20 Beta 1· LMArena data1,450.3
58Grok 4.5· LMArena data1,450.2
59MiMo V2.5 Pro· LMArena data1,449.6
60Doubao Seed 2.0 Pro· LMArena data1,449.4
61DeepSeek-V4-Pro PreviewHigh· LMArena data1,448.6
62Claude Opus 4.520251101· LMArena data1,447.2
63Claude Sonnet 5High· LMArena data1,446.6
64Gemini 3.5 Flash Lite· LMArena data1,445.6
65Qwen3.6 Max Preview· LMArena data1,445.4
66Kimi K2.6· LMArena data1,445
67DeepSeek V4 ProHigh, 20260813· LMArena data1,444.7
68ERNIE 5.00110· LMArena data1,444.6
69GPT-6 AstraMax· LMArena data1,443
70GPT-5.2 Chat20260210· LMArena data1,439.7
71Claude Sonnet 4.6· LMArena data1,438.9
72GPT-6.2 AstraMax· LMArena data1,437.1
73Claude Sonnet 4.520250929· LMArena data1,436.8
74GLM 5· LMArena data1,436.7
75DeepSeek V3.1 TerminusThinking· LMArena data1,436.6
76Claude Opus 4.5High, 32K, 20251101· LMArena data1,436.3
77ERNIE-5.0-Preview1203· LMArena data1,436.2
78GPT 5.1High· LMArena data1,435.2
79Kimi K2.5Thinking· LMArena data1,434.5
80Inkling FP4· LMArena data1,434.2
81Gemma 4 26B A4B· LMArena data1,434
82Qwen3.6 Plus· LMArena data1,434
83Grok 4.1Thinking· LMArena data1,434
84Hy3 (Deep Infra)· LMArena data1,433.8
85GPT-5.5 Instant· LMArena data1,432.6
86Step 5 Preview· LMArena data1,431.8
87Qwen3.5 397B A17B· LMArena data1,431.5
88ChatGPT-4o Latest20250326· LMArena data1,429.3
89Qwen3 Max Preview· LMArena data1,428.7
90Claude Sonnet 4.5High, 32K, 20250929· LMArena data1,427.6
91MiMo V2 Pro· LMArena data1,427.3
92GPT-5.6 LunaxHigh· LMArena data1,426.8
93Grok 4.1· LMArena data1,426.6
94DeepSeek V4 Flash· LMArena data1,426.2
95GPT 5.1· LMArena data1,426.2
96DeepSeek-V4-Flash PreviewHigh· LMArena data1,426
97GLM 4.7· LMArena data1,425
98DeepSeek V3.1 Terminus· LMArena data1,423.7
99DeepSeek V3.2· LMArena data1,423.2
100DeepSeek R10528· LMArena data1,422.4

Showing 51–100 of 387 results.

Data credits: LMArena, CC BY 4.0 (licence)