Compare / Claude Haiku 4.5 vs grok-3

Claude Haiku 4.5 vs grok-3

5 results from 2 sources that measured both models. Each row is in the source's own units; there is no overall winner.

Claude Haiku 4.5
Anthropic · profile
grok-3
xAI · profile
Results compared
5
Shared sources
2
BenchmarkMetricClaude Haiku 4.5grok-3
UGI LeaderboardreportedRequested-length error% off the requested word count 6.0%16.0%
UGI LeaderboardreportedStyle adherencescore from 0 to 1 0.390.36
UGI LeaderboardreportedWriting scorescore out of 100 52.258.7
Vectara Hallucination LeaderboardreportedAnswer rate% of documents 99.5%93.0%
Vectara Hallucination LeaderboardreportedHallucination rate% of summaries 9.8%5.8%

Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.