Publishers

23 labs with at least one model holding a board position, ordered by their best-ranked model.

• Anthropic

  4 ranked language models

  Language #1
  Best language model

  Claude Fable 5Measured · Index 92.3

  Output $5.00 – $50.00 / 1M
• Qwen

  19 ranked language models · 16 open

  Language #3
  Best language model

  Qwen3.8 MaxMeasured · Index 88.9

  Output $0.193 – $6.00 / 1M
• Moonshot AI

  3 ranked language models

  Language #4
  Best language model

  Kimi K3Measured · Index 88.7

  Output $2.50 – $15.00 / 1M
• OpenAI

  15 ranked language models · 3 open

  Language #5
  Best language model

  GPT-5.6 SolMeasured · Index 88.4

  Output $0.13 – $180 / 1M
• Meta

  7 ranked language models

  Language #7
  Best language model

  Muse Spark 1.2Measured · Index 87.4

  Output $0.08 – $4.25 / 1M
• Google

  9 ranked language models · 2 open

  Language #8
  Best language model

  Gemini 3.1 Pro PreviewMeasured · Index 86.9

  Output $0.10 – $12.00 / 1M
• xAI

  3 ranked language models

  Language #10
  Best language model

  Grok 4.20 Multi-AgentMeasured · Index 84.9

  Output $2.00 – $6.00 / 1M
• Z.ai

  5 ranked language models · 4 open

  Language #14
  Best language model

  GLM 5.2Measured · Index 83.3

  Output $0.40 – $4.00 / 1M
• Xiaomi

  2 ranked language models · 2 open

  Language #15
  Best language model

  MiMo-V2.5-ProMeasured · Index 81.4

  Output $0.28 – $0.87 / 1M
• DeepSeek

  5 ranked language models · 5 open

  Language #16
  Best language model

  DeepSeek V4 ProMeasured · Index 79.3

  Output $0.18 – $2.15 / 1M
• MiniMax

  1 ranked language model

  Language #23
  Best language model

  MiniMax M3Measured · Index 75.2

  Output $1.20 / 1M
• Tencent

  1 ranked language model · 1 open

  Language #24
  Best language model

  Hy3Measured · Index 75.1

  Output $0.528 / 1M
• TM

  Thinking Machines

  2 ranked language models · 2 open

  Language #29
  Best language model

  InklingMeasured · Index 73.3

  Output $1.20 – $4.05 / 1M
• Mistral AI

  4 ranked language models · 2 open

  Language #30
  Best language model

  Mistral Medium 3.5Measured · Index 72.0

  Output $0.25 – $7.50 / 1M
• NVIDIA

  3 ranked language models

  Language #31
  Best language model

  NVIDIA Nemotron 3 Ultra 550B A55B NVFP4Measured · Index 71.3

  Output $0.20 – $2.40 / 1M
• Arcee AI

  1 ranked language model

  Language #53
  Best language model

  Trinity Large ThinkingMeasured · Index 55.3

  Output $0.85 / 1M
• Cohere

  3 ranked language models

  Language #56
  Best language model

  Command AMeasured · Index 52.7

  Output $0.60 – $10.00 / 1M
• Amazon

  3 ranked language models

  Language #61
  Best language model

  Nova 2 LiteMeasured · Index 48.0

  Output $0.14 – $3.20 / 1M
• Inception Labs

  1 ranked language model

  Language #62
  Best language model

  Mercury 2Measured · Index 47.4

  Output $0.75 / 1M
• AI

  Allen Institute

  1 ranked language model · 1 open

  Language #75
  Best language model

  Olmo 3 32B ThinkMeasured · Index 37.5

  Output $0.50 / 1M
• IBM Granite

  1 ranked language model · 1 open

  Language #78
  Best language model

  Granite 4.1 8BMeasured · Index 36.1

  Output $0.10 / 1M
• Microsoft

  1 ranked language model · 1 open

  Language #81
  Best language model

  Phi 4Measured · Index 30.3

  Output $0.14 / 1M
• Qwen

  1 ranked language model · 1 open

  Language #91
  Best language model

  Qwen3 4B Instruct 2507Measured · Index 19.8