Dataset Viewer
Auto-converted to Parquet Duplicate
model_name
stringlengths
7
29
organization
stringclasses
9 values
license
stringclasses
4 values
rating
float64
1.4k
1.51k
rating_lower
float64
1.39k
1.5k
rating_upper
float64
1.41k
1.51k
variance
float64
6.72
84.6
vote_count
int64
1.09k
52.8k
rank
int64
1
32
category
stringclasses
1 value
leaderboard_publish_date
stringdate
2026-07-02 00:00:00
2026-07-02 00:00:00
claude-opus-4-6-thinking
anthropic
Proprietary
1,505.484116
1,498.44466
1,512.523572
12.899772
23,132
1
overall
2026-07-02
claude-opus-4-6
anthropic
Proprietary
1,504.734973
1,498.645897
1,510.824049
9.651763
35,514
2
overall
2026-07-02
claude-opus-4-7
anthropic
Proprietary
1,500.192383
1,493.592717
1,506.792049
11.338294
17,103
3
overall
2026-07-02
claude-opus-4-7-thinking
anthropic
Proprietary
1,499.617796
1,493.034951
1,506.20064
11.28057
16,879
4
overall
2026-07-02
claude-fable-5
anthropic
Proprietary
1,496.810635
1,485.060158
1,508.561112
35.943041
2,545
5
overall
2026-07-02
claude-sonnet-4-6
anthropic
Proprietary
1,487.590009
1,481.77002
1,493.409999
8.817555
52,833
6
overall
2026-07-02
gpt-5.5-high
openai
Proprietary
1,486.870328
1,480.249789
1,493.490867
11.410127
14,781
7
overall
2026-07-02
gpt-5.5
openai
Proprietary
1,481.367056
1,474.865376
1,487.868736
11.004113
15,246
8
overall
2026-07-02
gemini-3.5-flash
google
Proprietary
1,480.961414
1,465.047619
1,496.875208
65.925176
1,333
9
overall
2026-07-02
claude-opus-4-8-thinking
anthropic
Proprietary
1,476.643362
1,468.466612
1,484.820111
17.404645
6,376
10
overall
2026-07-02
claude-sonnet-5-thinking
anthropic
Proprietary
1,476.138216
1,459.483292
1,492.79314
72.208638
1,267
11
overall
2026-07-02
gpt-5.4
openai
Proprietary
1,474.875266
1,468.750123
1,481.000408
9.766439
27,648
12
overall
2026-07-02
claude-opus-4-8
anthropic
Proprietary
1,468.258739
1,459.7034
1,476.814078
19.053654
6,117
13
overall
2026-07-02
claude-opus-4-5-20251101
anthropic
Proprietary
1,461.285374
1,451.097507
1,471.47324
27.019064
7,987
14
overall
2026-07-02
kimi-k2.6
moonshot
Modified MIT
1,453.632722
1,446.035356
1,461.230087
15.025533
10,427
15
overall
2026-07-02
claude-sonnet-4-5-20250929
anthropic
Proprietary
1,446.909827
1,440.612009
1,453.207644
10.324855
26,859
16
overall
2026-07-02
muse-spark
meta
Proprietary
1,443.772893
1,425.740253
1,461.805533
84.649121
1,089
17
overall
2026-07-02
gemini-3.1-pro-preview
google
Proprietary
1,440.815999
1,435.735347
1,445.896651
6.719589
42,482
18
overall
2026-07-02
qwen3.7-plus
alibaba
Proprietary
1,439.197948
1,426.634254
1,451.761643
41.090227
2,051
19
overall
2026-07-02
minimax-m3
minimax
MiniMax Community License
1,435.36359
1,427.215404
1,443.511777
17.283263
6,367
20
overall
2026-07-02
gemini-3-pro
google
Proprietary
1,433.485079
1,424.803212
1,442.166946
19.621402
10,752
21
overall
2026-07-02
kimi-k2.5-thinking
moonshot
Modified MIT
1,429.173775
1,422.503742
1,435.843808
11.581366
18,646
22
overall
2026-07-02
gemma-4-31b
google
Apache 2.0
1,423.153676
1,414.937933
1,431.369418
17.571041
9,512
23
overall
2026-07-02
gemini-2.5-pro
google
Proprietary
1,421.33832
1,415.270179
1,427.40646
9.585507
25,094
24
overall
2026-07-02
claude-haiku-4-5-20251001
anthropic
Proprietary
1,420.789334
1,414.746507
1,426.832162
9.505702
29,105
25
overall
2026-07-02
grok-4.20-beta-0309-reasoning
xai
Proprietary
1,415.745327
1,408.780108
1,422.710547
12.62913
16,701
26
overall
2026-07-02
glm-5v-turbo
zai
Proprietary
1,415.627167
1,405.134871
1,426.119464
28.657936
3,334
27
overall
2026-07-02
gemini-3-flash
google
Proprietary
1,412.629834
1,403.492372
1,421.767296
21.734767
7,181
28
overall
2026-07-02
gpt-5.2-high
openai
Proprietary
1,405.075895
1,395.782757
1,414.369033
22.481672
7,091
29
overall
2026-07-02
gpt-5.5-instant
openai
Proprietary
1,402.959222
1,395.11374
1,410.804704
16.022973
8,536
30
overall
2026-07-02
gpt-5.2
openai
Proprietary
1,401.142429
1,395.432133
1,406.852725
8.488307
28,283
31
overall
2026-07-02
gpt-5.1
openai
Proprietary
1,400.981267
1,391.76062
1,410.201914
22.132303
8,247
32
overall
2026-07-02

Arena Leaderboard Dataset

Historical snapshots of the Arena leaderboard.

Usage

from datasets import load_dataset

# Load all historical text style control data
ds = load_dataset("lmarena-ai/leaderboard-dataset", "text_style_control", split="full")

# Load the current text style control leaderboard
ds = load_dataset("lmarena-ai/leaderboard-dataset", "text_style_control", split="latest")

# Filter to overall category
ds = load_dataset("lmarena-ai/leaderboard-dataset", "text_style_control", split="latest",
    filters=[("category", "==", "overall")]
)

# Track a specific model's rating over time
ds = load_dataset("lmarena-ai/leaderboard-dataset", "text_style_control", split="full",
    filters=[("category", "==", "overall"), ("model_name", "==", "gpt-4o-2024-05-13")],
    columns=["model_name", "rating", "rank", "leaderboard_publish_date"]
)

# Load raw (non-style-controlled) text ratings
ds = load_dataset("lmarena-ai/leaderboard-dataset", "text", split="full")

Subsets

Each arena is a separate subset, arenas with style control have an additional subset with a _style_control suffix .

  • text, text_style_control
  • vision, vision_style_control
  • search, search_style_control
  • document, document_style_control
  • webdev (Code Arena)
  • text_to_image
  • image_edit
  • text_to_video
  • image_to_video
  • video_edit

Splits

  • full: All historically published leaderboards
  • latest: Only the most recently published leaderboards

Notes

  • On January 9, 2024, the rating system was updated from Elo to Bradley-Terry.
  • On May 16, 2025, style control was made the default for text and vision arenas and the offset was adjusted to put the style control leaderboard on the same rating scale as non style control.
  • On July 23, 2025, frequency-based re-weighting was implemented.
  • Search and Webdev Arena data starts from their releases on the arena.ai (then lmarena.ai) on August 7 and November 12 2025 respectively.

For a complete record of all leaderboard methodology changes, see the Leaderboard Changelog.

Schema

Column Type Description
model_name string Model identifier
organization string Model creator/organization
license string Model license
rating float Arena Score
rating_lower float Lower confidence bound
rating_upper float Upper confidence bound
variance float Rating variance
vote_count int Number of battles for this model
rank int Rank within this leaderboard
category string Leaderboard category (e.g., overall, coding, math)
leaderboard_publish_date string Date that this score was published (YYYY-MM-DD)
Downloads last month
41