AI model rankings

New AI models

Every new model as soon as it's tested, newest first. Language, image, video and voice, with how each one ranks against the rest.

4 new this week

Yesterday
- MistralyesterdayLanguageNew
  Introducing Mistral Large 4
  Scores 26 on intelligence (#55 of 89), $1.05 per 1M tokens, about 127 words per second.
  2 versions:HighMedium
Wednesday, Oct 28
- Google2 days agoLanguageNew
  Introducing Gemini 4 Carbon
  Scores 59 on intelligence (#1 of 89), $8.00 per 1M tokens, about 66 words per second.
Monday, Oct 26
- OpenAI4 days agoLanguageNew
  Introducing GPT-6.1 Terra
  Scores 51 on intelligence (#12 of 89), $2.00 per 1M tokens, about 68 words per second.
  6 versions:MaxXhighHighMediumNon-reasoningLow
October 23, 2026
- KimiOct 23LanguageNew
  Introducing Kimi K3.1 Pro
  Scores 46 on intelligence (#22 of 89), $4.00 per 1M tokens, about 22 words per second.
October 22, 2026
- InclusionAIOct 22Language
  Introducing Ling 3.1 Flash 1019
  Scores 30 on intelligence (#48 of 89), $0.45 per 1M tokens.
October 19, 2026
- MiniMaxOct 19Language
  Introducing MiniMax-M3.1-Flash
  Scores 29 on intelligence (#50 of 89), $0.26 per 1M tokens, about 85 words per second.
October 17, 2026
- DeepSeekOct 17Language
  Introducing DeepSeek V4.1 Flash 1017
  Scores 33 on intelligence (#39 of 89), $0.53 per 1M tokens, about 119 words per second.
  2 versions:MaxNon-reasoning
October 16, 2026
- ReflectionOct 16Language
  Introducing Beam 1.0
  Scores 18 on intelligence (#72 of 89), $0.60 per 1M tokens, about 233 words per second.
October 15, 2026
- MetaOct 15Language
  Introducing Muse Spark Open
  Scores 25 on intelligence (#58 of 89), $2.00 per 1M tokens, about 105 words per second.
- NVIDIAOct 15Language
  Introducing Nemotron 3.5 Super VL
  Scores 14 on intelligence (#76 of 89), $0.31 per 1M tokens, about 313 words per second.
October 12, 2026
- MetaOct 12Language
  Introducing Muse Wonder Full
  Scores 50 on intelligence (#15 of 89).
- MetaOct 12Language
  Introducing Muse Wonder
  Scores 43 on intelligence (#24 of 89), $8.00 per 1M tokens, about 112 words per second.
October 7, 2026
- AnthropicOct 7Language
  Introducing Claude Haiku 5.5
  Scores 36 on intelligence (#32 of 89), $0.20 per 1M tokens, about 177 words per second.
  5 versions:MaxXhighHighMediumLow
October 6, 2026
- MistralOct 6Language
  Introducing Mistral Large 4 Preview
  Scores 25 on intelligence (#57 of 89), $1.05 per 1M tokens, about 84 words per second.
October 2026 (exact day unknown)

Benchmark data from Artificial Analysis, an independent benchmarking company. Rankings refresh every few hours.

Updated Oct 30, 2026, 05:43 PM UTC