Head-to-head token pricing, context window limits, and blended inference cost comparison between Claude Haiku 5.5 (Anthropic) and Gemini 3.8 Flash (Google).
Gemini 3.8 Flash costs 8.2x more than Claude Haiku 5.5 on a 3:1 blended token workload. Claude Haiku 5.5 has a context window of 1000k tokens, while Gemini 3.8 Flash provides 1049k tokens.