Anthropic Launches Haiku 5.5, Its Cheapest and Fastest Claude Model

The pricing gap between Anthropic's flagship and its budget tier is now the story, not the model itself. Positioning Haiku 5.5 explicitly at high-volume, low-complexity tasks signals where the vendor expects AI inference margins to compress first — and it pressures rival labs to match on cost per token rather than capability benchmarks. The 15-day gap from Opus 5.5 suggests a deliberate cadence of premium launches followed by volume plays. Whether enterprise buyers actually shift live support and summarization workloads down to the cheaper tier, or keep routing them to larger models out of caution, is the open question.