OpenGPT
Search⌘K
Public ranking sources

LiveBench AI model ranking data

OpenGPT preserves 30 published LiveBench records in their original source scope. It does not rerun or blend this ranking with another benchmark.

Published records
30
Ranking views
Objective tasks · Cost per score point · Open-weight models
Source data date
Retrieved

Evidence scope

LiveBench-2026-06-25

62 published model configurations across twenty-three objective tasks and seven categories; questions refresh every six months.

How OpenGPT handles this evidence

OpenGPT stores the publisher's labels, positions, configurations, dates, and values within the same published scope. It does not convert missing values to zero or merge this source into a cross-source score.

Comparison boundaries

  • Ranks and values are comparable only inside the named source release and ranking scope.
  • A source-listed configuration is linked to an official model ID only when identity evidence has been reviewed.
  • Published rankings are screening signals; production choice still requires testing the exact version on real work.

LiveBench-2026-06-25

Showing 18 representative rows from 30 published records. Follow a row to inspect its exact source identity and mapping evidence.

  1. Claude Fable 5.1 Max EffortAnthropic · Objective tasks#183.41Published benchmark configuration
  2. Claude 5.5 Opus Thinking Max EffortAnthropic · Objective tasks#283.22Published benchmark configuration
  3. Claude Fable 5 Max EffortAnthropic · Objective tasks#382.97Published benchmark configuration
  4. GPT-6 Astra Max EffortOpenAI · Objective tasks#482.16Published benchmark configuration
  5. GPT-6.1 Sol Max EffortOpenAI · Objective tasks#581.62Published benchmark configuration
  6. Muse Spark 1.3 xHigh EffortMeta · Objective tasks#681.59Published benchmark configuration
  7. Claude Haiku 5.5 xHigh EffortAnthropic · Cost per score point#1$0.0188Published benchmark configuration
  8. Grok Build 0.1xAI · Cost per score point#2$0.0240Published benchmark configuration
  9. GPT-6 Luna Max EffortOpenAI · Cost per score point#3$0.0263Published benchmark configuration
  10. DeepSeek V4.1 Flash Max EffortDeepSeek · Cost per score point#4$0.0292Published benchmark configuration
  11. GLM-5.3 FlashZ.AI · Cost per score point#5$0.0305Published benchmark configuration
  12. Nemotron 3.5 Lightning 30B A3BNVIDIA · Cost per score point#6$0.0399Published benchmark configuration
  13. DeepSeek V4.1 Flash Max EffortDeepSeek · Open-weight models#181.11Open weights
  14. Kimi K3Moonshot AI · Open-weight models#279.19Open weights
  15. Qwen 3.8 MaxAlibaba · Open-weight models#378.46Open weights
  16. DeepSeek V4 Pro 0813DeepSeek · Open-weight models#477.44Open weights
  17. DeepSeek V4 Flash Vision ExpDeepSeek · Open-weight models#576.76Open weights
  18. Qwen 3.8 Flash NextAlibaba · Open-weight models#676.19Open weights