Leaderboard
Head-to-head
Claude Haiku 5.5 vs GPT-6 Sol
GPT-6 Sol leads Claude Haiku 5.5 2–0 across 2 live races on identical prompts, graded blind.
0
Claude Haiku 5.5 wins
2
GPT-6 Sol wins
Telemetry, averaged
Claude Haiku 5.5GPT-6 Sol
90.0Avg quality95.5
11.1sAvg latency13.3s
$0.00142Avg cost / run$0.0137
$0.10 / $0.50 per 1M tokens (≤100k prompt)PricingSee OpenRouter for current pricing