OpenAI is aggressively responding to the intensifying market for large language models by implementing substantial price cuts across its GPT-5.6 series. The company announced an 80% reduction in the cost for its entry-level Luna model, which now carries a combined input and output price of $1.40 per million tokens. Additionally, the mid-tier Terra model has seen a 20% price adjustment, bringing its total cost to $14 per million tokens. These changes are part of a broader industry shift where major providers are increasingly prioritizing cost efficiency and inference speed to attract developers.

According to VentureBeat, these strategic pricing maneuvers arrive on the heels of competitive product launches from Google and Anthropic. Google recently expanded its portfolio with the Gemini 3.6 Flash and Flash-Lite models, designed specifically for rapid, cost-effective agentic workflows. Meanwhile, Anthropic has sustained its performance-focused strategy, releasing the updated Claude Opus 5 at a price point consistent with its predecessor. OpenAI's decision to undercut rivals while simultaneously introducing a 'Sol Fast' mode suggests a bifurcated strategy: maintaining high-end performance for premium users while drastically lowering the barrier to entry for lighter, speed-oriented tasks.

The new pricing structure positions Luna as a direct challenger to lower-cost market entrants. While the standard pricing for the flagship GPT-5.6 Sol model remains unchanged at $35 per million tokens, the newly introduced Fast mode offers a throughput boost of 2.5 times the standard rate for double the cost. By diversifying its service tiers, OpenAI aims to capture a wider share of the developer ecosystem, challenging Google’s aggressive pricing models while providing existing power users with a more performant option for latency-sensitive applications.