Skip to content

Every board we track

29 AI leaderboards, who is #1 on each, and when each one last published.
Text Arena
LMArena
Chat
1Claude Fable 51,506
2Claude Opus 4.61,505
3Claude Opus 4.71,502
Published Sep 13, 2026100 models
MultiChallenge
Scale AI
Chat
1Muse Spark75.5%
2Muse Spark 1.175.3%
3Gemini 3.1 Pro Preview71.4%
Published Sep 24, 202629 models
AA Intelligence
Artificial Analysis
Reasoning
1Claude Opus 5.557.6
2Claude Fable 5.153.4
3GPT-6 Astra52.7
Published Sep 24, 2026100 models
HLE
Scale AI and CAIS
Reasoning
1GPT-6 Astra54.8%
2Claude Fable 5.146.5%
3Gemini 3.1 Pro Preview46.4%
Published Sep 24, 202642 models
GPQA Diamond
Epoch AI
Reasoning
1GPT-6 Astra95.8%
2Gemini 3.8 Flash95.4%
3Gemini 3.7 Flash94.8%
Published Sep 2, 2026100 models
Epoch ECI
Epoch AI
Reasoning
1GPT-6 Astra166.6
2Claude Fable 5.1165.0
3Claude Fable 5163.6
Published Sep 24, 2026100 models
ARC-AGI-2
ARC Prize
Reasoning
1GPT-6 Astra95.0%
2Claude Opus 5.593.3%
3GPT-5.6 Sol92.5%
Published Sep 22, 202678 models
LiveBench
LiveBench
Reasoning
1Claude Fable 5.183.4%
2Claude Opus 5.583.2%
3Claude Fable 583.0%
Published Jun 25, 202659 models
MMLU-Pro
TIGER-Lab
Reasoning
1Gemini 3.1 Pro Preview91.2%
2Gemini 3 Pro90.1%
3o189.3%
Stale since 2026-03-11100 models
WebDev Arena
LMArena
Coding
1Claude Opus 5.51,818
2GPT-6 Astra1,792
3Claude Fable 5.11,755
Published Sep 23, 2026100 models
SWE-Bench Pro
Scale AI
Coding
1Claude Opus 598.0%
2Claude Fable 5.192.2%
3GPT-6 Astra90.2%
Published Sep 24, 202611 models
SWE-bench Verified
SWE-bench
Coding
1Claude Opus 4.576.8%
2Gemini 3 Flash75.8%
2MiniMax-M2.575.8%
Stale since 2026-02-2639 models
Aider Polyglot
Aider
Coding
1GPT-588.0%
2o3-pro84.9%
3Gemini 2.5 Pro (Jun 2025)83.1%
Stale since 2025-10-0350 models
BFCL
UC Berkeley Gorilla
Agents and tools
1Claude Opus 4.577.5%
2Claude Sonnet 4.573.2%
3Gemini 3 Pro72.5%
Published Apr 12, 202681 models
MCP Atlas
Scale AI
Agents and tools
1Muse Spark 1.188.1%
2Claude Fable 5.187.2%
3Claude Opus 585.8%
Published Sep 23, 202634 models
Search Arena
LMArena
Agents and tools
1GPT-5.6 Sol1,257
2Claude Opus 4.61,253
3GPT-5.51,242
Published Aug 24, 202633 models
Vision Arena
LMArena
Vision
1Claude Fable 51,310
2Qwen3.8 Max (0902)1,302
3Claude Opus 4.71,301
Published Sep 13, 2026100 models
Document Arena
LMArena
Vision
1Claude Opus 51,516
2Claude Fable 5.11,513
3Claude Opus 4.61,507
Published Sep 13, 202638 models
Text-to-Image
LMArena
Image
1GPT Image 2.5 Sunburst1,423
2GPT Image 2.5 Flare1,401
3GPT Image 21,381
Published Sep 21, 202675 models
Image Edit
LMArena
Image
1GPT Image 2.5 Sunburst1,526
2GPT Image 2.5 Flare1,482
3GPT Image 21,461
Published Sep 21, 202653 models
Text-to-Video
LMArena
Video
1Gemini Omni 1.1 Flash1,516
2Gemini Omni Flash1,513
3Flux 3 Video1,493
Published Sep 21, 202648 models
Image-to-Video
LMArena
Video
1MiniMax H31,495
2Gemini Omni 1.1 Flash1,488
3Wan3.01,480
Published Sep 21, 202648 models
Speech Arena
Artificial Analysis
Speech
1Sonic 3.61,273
2Gemini 3.8 Flash TTS1,260
3Qwen Audio 3.0 TTS Plus1,259
Published Sep 24, 202691 models
Open ASR
Hugging Face
Speech
1scribe v2 pro (zoom)3.59 WER
2azure-speech-07-2026 (microsoft)3.81 WER
3multilingual (modulate)3.84 WER
Published Sep 22, 202667 models
MTEB
MTEB
Embeddings
1harrier-oss-v1-27b (microsoft)74.3%
2KaLM-Embedding-Gemma3-12B-2511 (tencent)72.3%
3llama-embed-nemotron-8b (nvidia)69.5%
Published Sep 22, 202688 models
AA Speed
Artificial Analysis
Speed
1Gemini 3.8 Flash291 tok/s
2DeepSeek V4.1 Flash232 tok/s
3Muse Spark 1.3228 tok/s
Published Sep 24, 202624 models
AA Price
Artificial Analysis
Price
1Agnes 3.0 Flash$0.07
2Agnes 2.5 Pro Beta$0.15
3GPT-6 Luna$0.20
Published Sep 24, 202627 models
OpenRouter
OpenRouter
Usage
1GLM 5.3 Flash19.0T tok
2DeepSeek V4.1 Flash18.4T tok
3Hy4 preview13.0T tok
Published Sep 23, 202619 models
Open LLM LB
Hugging Face
Reasoning

Retired 2025-03-20. Listed for reference.

Retired

Scores as published by each board on the capture date. Model names and logos belong to their owners; logos via logo.dev.

Weekly: the AI leaderboards with a new #1, Saturday mornings.