Anthropic, OpenAI Release Cheaper AI Models Amid Stiff Competition

Anthropic and OpenAI have launched cheaper AI models amid growing competition from open-weight competitors. The new models expand their lineups as enterprises seek affordable options. Anthropic introduced Claude Opus 5.5, while OpenAI unveiled GPT-6 Sol and GPT-6 Luna. The low cost AI models are designed to support everyday tasks in businesses. 

What Tasks Can OpenAI’s Cheaper GPT-6 Models Perform? 

The newly launched AI models expand OpenAI’s GPT-6 lineup beyond the recently launched GPT-6 Astra. GPT-6 Sol is a tier below Astra and is designed to handle more complex workloads such as coding and agentic tasks. 

GPT-6 Luna, on the other hand, handles high-volume, routine workloads such as summarizing documents and extracting information. OpenAI priced APIs for the two models at 50% lower than the promotional rates for the GPT-5.6 models. The AI firm attributed the price reduction to optimizations in inference and caching, saying it is passing the savings on to customers.

Claude Opus 5.5 Comes With Better Token-Efficiency

Anthropic said its latest model, Claude Opus 5.5, is more token-efficient. It costs about 40% less than Opus 5 for typical workloads and is about 30% faster. The company attributed the lower cost to lower compute requirements. 

“One of the things we’re continuing to innovate on is how to make that thinking, how to make the answering more efficient, so it uses fewer tokens depending on your effort setting,” Anthropic’s Head of Product Management, Research and Labs, Dianne Penn, said, as cited by CNBC. 

Anthropic priced input and output tokens at $4 and $20 per million, respectively, which is about 20% lower than Opus 5. The company said cache reads, which constitute a large portion of coding and agentic work costs, will cost $0.20 per million tokens. This is roughly 60% less than Opus 5. The AI firm said the new model performance is comparable to Fable 5.1 on most tasks. 

Why Cheaper AI Models from OpenAI and Anthropic Matter 

The new releases are designed to meet customers’ needs for more cost-effective AI models.  Anthropic and OpenAI are facing stiff competition from Chinese labs offering open-weight AI models at lower costs. 

Recently, Moonshot AI launched Kimi K3, which the company claimed performed better than OpenAI’s GPT-5.5 and Anthropic’s Claude Opus 4.8 in benchmarks like general agents and coding. Moonshot priced Kimi K3 at $15 per 1 million output tokens, about $0.94 per task. This is slightly lower than GPT-5.6 Sol’s $1.04, and half of Claude Opus 4.8’s $1.80.

The cheaper models also launched soon after Anthropic’s Dario Amodei and OpenAI’s Sam Altman called for a slowdown in frontier AI development as discussions about model safety continue. EU President Ursula von der Leyen backed the call and committed to invite AI firms to discuss ways to support ongoing industry efforts to pace the frontier.