Anthropic expands AI model suite with low-cost, high-speed Haiku 5.5 launch

Investing.com -- Anthropic (NASDAQ: ANTP) announced the launch of Claude Haiku 5.5 on Wednesday, introducing a substantially faster and cheaper small-format artificial intelligence model aimed at high-volume enterprise workflows and subagent task execution. The release reflects an aggressive expansion of the company's tier-priced model matrix as competitive pressure intensifies across the generative AI ecosystem.

The new iteration delivers roughly a 75% average cost reduction compared to its predecessor, Haiku 4.5, while offering up to 90% savings on standard prompt loads under 100,000 tokens. To further drive adoption among developers and enterprise clients, Anthropic coupled the deployment with a 50% price cut on cache reads for its flagship Sonnet 5.5 model, cutting overall compute expenses for complex agentic tasks by approximately 20%.

Early customer testing highlights significant operational efficiencies, with software management platform Asana reporting over a 30% reduction in task completion latency and up to 2.5 times faster inference speeds. These operational gains position Haiku 5.5 as a tailored solution for latency-sensitive applications, including automated triage, real-time customer support, and high-frequency database queries.

Benchmarking data underscores dramatic performance jumps across technical evaluations, with Haiku 5.5 achieving a 72.4% success rate on the OSWorld 2.1 computer-use benchmark, up from 15.7% in the prior generation. The model also marks the first in the Haiku class to feature adjustable reasoning effort settings, enabling developers to dynamically balance cost parameters against intelligence requirements based on specific workload needs.

Beyond API performance enhancements, the deployment introduces a structured incentive program to deepen developer engagement across Anthropic's platform ecosystem. Premium Max and Team subscription tiers will now receive monthly API credits ranging from $100 to $500, directly subsidizing cross-model experimentation and real-time application development.

The strategic repositioning of the Haiku framework highlights Anthropic's focus on capturing high-volume enterprise workloads while reserving its larger Sonnet and Opus models for heavy-duty coding and complex reasoning tasks. By lowering entry costs and broadening access across major cloud platforms, including AWS, Google Cloud, and Microsoft Azure, the company reinforces its broader platform strategy against key industry rivals.

Original Article

You May Also Be Interested In

• OpenAI annualized revenue at $50bn, far below reports - FT
• Gap sees executive exit as catalyst to rebuild struggling Athleta unit - Bloomberg
• Haemonetics stock surges 17% on CSL Plasma supply deal update