Anthropic released Claude Opus 5 on Friday, positioning it as a cost-competitive alternative to its flagship Claude Fable 5 model. The new offering delivers nearly identical intelligence at half the cost, marking a shift in how AI vendors compete on price rather than raw capability alone.
Opus 5 launches across all Anthropic platforms immediately with pricing identical to its predecessor, Opus 4.8: $5 per million input tokens and $25 per million output tokens. The model becomes the default on Claude Max, Anthropic's paid tier.
The release reflects a maturing AI market. As large language models become commoditized, vendors increasingly focus on operational economics rather than marginal performance gains. Opus 5 targets enterprise workflows, coding tasks, and agent-based applications where cost per inference matters more than achieving state-of-the-art benchmarks.
Anthropic's strategy differs from competitors who launch new models at premium pricing. By matching Opus 4.8 pricing while improving capabilities, Anthropic captures price-sensitive enterprise customers without cannibalizing revenue from higher-tier offerings. Fable 5 remains available for users requiring absolute maximum performance.
The move reveals how AI commodification accelerates. Anthropic previously faced pressure from OpenAI's GPT-4o, which offered strong performance at lower costs. By releasing Opus 5 at existing price points rather than premium tiers, Anthropic removes pricing as a friction point for adoption.
Enterprise deployment drives this economics focus. Production AI systems run thousands or millions of inferences daily. A 50% cost reduction translates directly to operational savings. For coding tasks, where latency and minor quality differences matter less than throughput and cost, Opus 5 becomes the pragmatic choice.
Anthropic has not disclosed comparative benchmarks between Opus 5 and