GPT-6 Sol And Luna Debut With 50% Price Cuts, Stronger Caching, But Independent Tests Show Mixed Results

by
Alisa Davidson

Revealed: September 23, 2026 at 4:07 am Up to date: September 23, 2026 at 4:08 am

by Anastasiia O

Edited and fact-checked:
September 23, 2026 at 4:07 am

To enhance your local-language expertise, typically we make use of an auto-translation plugin. Please observe auto-translation is probably not correct, so learn original article for exact data.

OpenAI has formally launched two new fashions, GPT-6 Sol and GPT-6 Luna, increasing the GPT-6 household alongside the flagship GPT-6 Astra launched earlier this month. In response to the corporate, the brand new fashions ship near-Astra-level efficiency in skilled work, factuality, coding, and laptop use, whereas slicing API costs by 50% in comparison with the promotional pricing of the earlier GPT-5.6 era. The corporate attributes the price discount to enhancements in caching and inference infrastructure, the financial savings from which it says are being handed on to customers.

Beneath the brand new pricing, GPT-6 Sol prices $2 per million enter tokens and $10 per million output tokens, down from $4 and $20 respectively, whereas GPT-6 Luna is priced at $0.10 per million enter tokens and $0.50 per million output tokens, down from $0.20 and $1.20. OpenAI positions Astra as its uncompromising finest mannequin for probably the most demanding tasks, whereas Sol and Luna are designed to make superior AI sensible for higher-volume, on a regular basis workloads.

Benchmark Outcomes and Technical Enhancements

On AutomationBench, which checks brokers throughout 47 enterprise instruments in gross sales, advertising and marketing, operations, assist, finance, and HR, GPT-6 Sol at xhigh effort scored 33.2% at $0.27 per activity, outperforming Claude Opus 5 at max effort (26.9%) at roughly 9% of its value per activity. On Brokers’ Final Examination, Sol at max effort reached 56.4%, exceeding Claude Opus 5’s highest rating at about 60% decrease value. For coding, GPT-6 Sol scored 68.8% on DeepSWE v1.1, inside 1.1 proportion factors of Claude Fable 5’s finest end result at roughly 80% decrease value, whereas Luna’s 66.6% was similar to Opus 5 and Fable 5 at medium effort at 93–96% decrease value. On OSWorld 2.0, Sol matched Claude Opus 5 (60.5% versus 60.3%) at about one-fifth of the price.

OpenAI additionally reviews halved factual error charges for Sol relative to its predecessor on its inside analysis of de-identified conversations the place customers flagged errors, with Luna at greater effort matching GPT-5.6 Sol’s reliability at roughly one-hundredth of the price. Alignment evaluations present each fashions bettering over their predecessors, together with decrease charges of deception in intentionally difficult coding situations.

Past pricing, OpenAI has improved immediate caching to boost cache hit charges by default, providing a 90% low cost on cached input-token reads, alongside a monitoring dashboard, diagnostics instruments, and controls that allow builders modify reasoning effort and power availability with out invalidating cached context. GitHub reviews these modifications have lower the share of immediate tokens requiring recent processing by greater than 50% throughout billions of requests.

GPT-6 Sol and Luna can be found beginning right this moment in ChatGPT Work and Codex for Plus, Professional, Enterprise, Enterprise, and Edu customers, with Luna additionally accessible to Free and Go customers by way of the desktop app; each are supplied within the API as gpt-6-sol and gpt-6-luna, with a gradual rollout all through the day.

Unbiased Evaluation Confirms Price Positive factors, Combined Outcomes

Third-party analysis by Synthetic Evaluation largely corroborates OpenAI’s effectivity claims whereas portray a extra nuanced image of the fashions’ capabilities. Working its Intelligence Index, the agency discovered that GPT-6 Sol at max effort prices roughly $1.06 per activity, roughly half the $1.99 of its predecessor, whereas Luna drops from $0.18 to $0.07 per activity — inserting each releases firmly on the cost-efficiency Pareto frontier. Notably, the financial savings stem completely from the worth lower, as each fashions devour barely extra output tokens per activity than their predecessors.

Efficiency, nonetheless, is a mixture of progress and regression. Within the Coding Agent Index, Sol improved by 2 factors to 57, with features in Terminal-Bench 4.0 and SWE-Atlas-QnA, whereas Luna misplaced 2 factors, falling behind in SWE-Atlas-QnA and DeepSWE v1.1. Hallucination charges fell sharply on the AA-Omniscience benchmark — from 92% to 60% for Sol — although this was partly achieved by declining to reply extra questions, which decreased Sol’s accuracy by 5 factors. Probably the most vital regressions appeared in knowledge-work evaluations: Sol dropped roughly 100 Elo factors in GDPval-AA v2.1 and Luna about 75, with handbook inspection attributing the declines to shorter deliverables that omit required rubric parts and decreased presentation high quality.

Alisa, a devoted journalist on the MPost, makes a speciality of crypto, AI, investments, and the expansive realm of Web3. With a eager eye for rising tendencies and applied sciences, she delivers complete protection to tell and have interaction readers within the ever-evolving panorama of digital finance.

• 
  
  
  

More articles