LMArena
Indexed results from LMArena. Each ranking retains its own metric and evaluation conditions.
LMArena Text Factuality Industry Life And Physical And Social Science
lmarena · LMArena · Arena rating · rating
- Google: Gemini 4 Argongoogle1,531.128
- Anthropic: Claude Opus 5anthropic1,529.293
- Google: Gemini 4 Hydrogengoogle1,519.083
- Anthropic: Claude Opus 5anthropic1,513.044
- Anthropic: Claude Fable 5anthropic1,512.503
- Google: Gemini 3.7 Flashgoogle1,510.019
- OpenAI: GPT-5.5openai1,506.319
- Google: Gemini 3.8 Flashgoogle1,506.148
- OpenAI: GPT-5.5openai1,505.764
- Google: Gemini 3.5 Flashgoogle1,503.152
- OpenAI: GPT-5.4openai1,499.114
- Google: Gemini 3.5 Flashgoogle1,495.288
LMArena Text Factuality Industry Mathematical
lmarena · LMArena · Arena rating · rating
- Anthropic: Claude Opus 5anthropic1,520.253
- Google: Gemini 4 Argongoogle1,511.215
- Anthropic: Claude Opus 5anthropic1,509.651
- OpenAI: GPT-5.4openai1,497.624
- Anthropic: Claude Fable 5anthropic1,494.976
- OpenAI: GPT-5.5openai1,492.216
- Google: Gemini 3.8 Flashgoogle1,491.261
- OpenAI: GPT-5.6 Solopenai1,489.41
- OpenAI: GPT-5.5openai1,484.585
- OpenAI: GPT-6 Astraopenai1,483.999
- Google: Gemini 3.5 Flashgoogle1,483.122
- OpenAI: GPT-5.6 Terraopenai1,482.591
LMArena Text Factuality Industry Medicine And Healthcare
lmarena · LMArena · Arena rating · rating
- Google: Gemini 4 Argongoogle1,535.172
- Anthropic: Claude Opus 5anthropic1,527.624
- Anthropic: Claude Opus 5anthropic1,516.797
- Google: Gemini 3.8 Flashgoogle1,505.53
- Anthropic: Claude Fable 5anthropic1,505.371
- Google: Gemini 3.5 Flashgoogle1,504.668
- OpenAI: GPT-5.5openai1,502.555
- Google: Gemini 3.7 Flashgoogle1,501.588
- OpenAI: GPT-5.4openai1,501.109
- Google: Gemini 2.5 Progoogle1,500.944
- OpenAI: GPT-5.5openai1,500.199
- Google: Gemini 3.5 Flashgoogle1,499.316
LMArena Text Factuality Industry Software And It Services
lmarena · LMArena · Arena rating · rating
- Google: Gemini 4 Carbongoogle1,546.598
- Google: Gemini 4 Argongoogle1,536.136
- Anthropic: Claude Opus 5anthropic1,527.02
- Google: Gemini 4 Hydrogengoogle1,526.09
- OpenAI: GPT-5.5openai1,522.451
- Anthropic: Claude Fable 5anthropic1,520.423
- OpenAI: GPT-5.4openai1,519.25
- Anthropic: Claude Opus 5anthropic1,518.616
- OpenAI: GPT-5.6 Solopenai1,517.542
- OpenAI: GPT-5.5openai1,516.539
- OpenAI: GPT-6 Astraopenai1,514.745
- OpenAI: GPT-6.1 Solopenai1,513.438
LMArena Text Factuality Industry Writing And Literature And Language
lmarena · LMArena · Arena rating · rating
- Google: Gemini 4 Carbongoogle1,522.42
- Google: Gemini 4 Argongoogle1,509.586
- Anthropic: Claude Fable 5anthropic1,489.568
- Google: Gemini 4 Hydrogengoogle1,485.581
- Anthropic: Claude Opus 5anthropic1,484.831
- Google: Gemini 3.5 Flashgoogle1,477.334
- Google: Gemini 3.7 Flashgoogle1,475.483
- Anthropic: Claude Opus 5anthropic1,474.465
- Google: Gemini 3.8 Flashgoogle1,473.416
- Google: Gemini 3.1 Pro Previewgoogle1,470.277
- OpenAI: GPT-5.5openai1,470.032
- OpenAI: GPT-5.5openai1,469.088
LMArena Text Factuality Instruction Following
lmarena · LMArena · Arena rating · rating
- Google: Gemini 4 Carbongoogle1,523.913
- Google: Gemini 4 Hydrogengoogle1,517.133
- Google: Gemini 4 Argongoogle1,515.068
- Anthropic: Claude Opus 5anthropic1,501.38
- Anthropic: Claude Opus 5anthropic1,494.702
- Anthropic: Claude Fable 5anthropic1,494.324
- OpenAI: GPT-5.6 Solopenai1,491.19
- OpenAI: GPT-5.5openai1,490.71
- OpenAI: GPT-5.5openai1,488.707
- OpenAI: GPT-5.4openai1,483.822
- OpenAI: GPT-6.1 Solopenai1,479.961
- Google: Gemini 3.7 Flashgoogle1,477.862
LMArena Text Factuality Japanese
lmarena · LMArena · Arena rating · rating
- Anthropic: Claude Opus 5anthropic1,518.382
- Google: Gemini 3.8 Flashgoogle1,513.839
- OpenAI: GPT-5.5openai1,511.158
- OpenAI: GPT-5.5openai1,492.766
- Anthropic: Claude Fable 5anthropic1,486.212
- OpenAI: GPT-5.4openai1,484.218
- Google: Gemini 3.1 Pro Previewgoogle1,482.074
- OpenAI: GPT-5.4openai1,469.707
- Google: Gemini 3.5 Flashgoogle1,465.552
- OpenAI: GPT-5.6 Terraopenai1,460.339
- kimi-k2.6moonshot1,444.805
- Anthropic: Claude Sonnet 5anthropic1,443.08
LMArena Text Factuality Korean
lmarena · LMArena · Arena rating · rating
- Google: Gemini 3.8 Flashgoogle1,503.155
- Anthropic: Claude Opus 5anthropic1,476.108
- Google: Gemini 3.1 Pro Previewgoogle1,459.833
- OpenAI: GPT-5.5openai1,455.558
- OpenAI: GPT-5.5openai1,446.597
- OpenAI: GPT-5.6 Solopenai1,445.022
- OpenAI: GPT-5.4openai1,438.827
- OpenAI: GPT-5.6 Terraopenai1,434.908
- google: Gemini 3 Flashgoogle1,422.065
- Google: Gemini 2.5 Progoogle1,416.449
- claude-opus-4-5@20251101anthropic1,416.199
- Anthropic: Claude Sonnet 4.5anthropic1,409.919
LMArena Text Factuality Longer Query
lmarena · LMArena · Arena rating · rating
- Google: Gemini 4 Carbongoogle1,538.768
- Google: Gemini 4 Argongoogle1,533.284
- Google: Gemini 4 Hydrogengoogle1,525.08
- Anthropic: Claude Opus 5anthropic1,506.627
- Anthropic: Claude Fable 5anthropic1,504.634
- OpenAI: GPT-5.5openai1,500.128
- Anthropic: Claude Opus 5anthropic1,498.457
- OpenAI: GPT-5.5openai1,497.67
- OpenAI: GPT-5.6 Solopenai1,494.119
- Google: Gemini 3.8 Flashgoogle1,493.773
- claude-opus-4-5@20251101anthropic1,491.555
- Google: Gemini 3.5 Flashgoogle1,491.4
LMArena Text Factuality Math
lmarena · LMArena · Arena rating · rating
- Anthropic: Claude Opus 5anthropic1,516.634
- Anthropic: Claude Opus 5anthropic1,515.56
- Anthropic: Claude Fable 5anthropic1,507.963
- Google: Gemini 3.8 Flashgoogle1,503.482
- Google: Gemini 4 Argongoogle1,501.016
- Google: Gemini 3.7 Flashgoogle1,495.36
- OpenAI: GPT-5.5openai1,494.104
- OpenAI: GPT-6 Astraopenai1,491.748
- Google: Gemini 3.5 Flashgoogle1,490.68
- Google: Gemini 3.6 Flashgoogle1,490.498
- OpenAI: GPT-5.4openai1,490.35
- OpenAI: GPT-5.6 Solopenai1,486.265
LMArena Text Factuality Multi Turn
lmarena · LMArena · Arena rating · rating
- Google: Gemini 4 Argongoogle1,532.06
- Anthropic: Claude Fable 5anthropic1,503.913
- OpenAI: GPT-5.4openai1,499.989
- Anthropic: Claude Opus 5anthropic1,496.592
- OpenAI: GPT-5.5openai1,493.135
- OpenAI: GPT-5.5openai1,491.602
- Google: Gemini 3.8 Flashgoogle1,489.901
- Google: Gemini 3.5 Flashgoogle1,489.155
- OpenAI: GPT-5.4openai1,489.015
- Google: Gemini 3.7 Flashgoogle1,486.14
- Anthropic: Claude Opus 5anthropic1,486.117
- Google: Gemini 3.6 Flashgoogle1,483.663
LMArena Text Factuality Non English
lmarena · LMArena · Arena rating · rating
- Google: Gemini 4 Carbongoogle1,511.708
- Google: Gemini 4 Argongoogle1,488.016
- Google: Gemini 4 Hydrogengoogle1,486.097
- Anthropic: Claude Opus 5anthropic1,479.408
- OpenAI: GPT-5.5openai1,478.419
- OpenAI: GPT-5.4openai1,475.334
- Anthropic: Claude Fable 5anthropic1,471.312
- OpenAI: GPT-5.5openai1,470.638
- Google: Gemini 3.5 Flashgoogle1,469.601
- Google: Gemini 3.7 Flashgoogle1,468.426
- Anthropic: Claude Opus 5anthropic1,467.126
- OpenAI: GPT-5.6 Solopenai1,466.781
Models
LoadingBenchmarks
Explore every benchmark
LoadingThese charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.