OpenAI Cuts Prices In Half For GPT‑6 Sol And Luna Without Changing Benchmark Scores
AIThis post was created with the assistance of artificial intelligence (AI).

🔍 Read the full analysis: OpenAI Cuts Prices In Half For GPT‑6 Sol And Luna Without Changing Benchmark Scores on ThorstenMeyerAI.com

Buying for a business?Offer from Amazon

Get business pricing on monitors, keyboards and dev gear

  • Business-only prices and quantity discounts
  • Tax-exempt purchasing
  • Multiple users, one account, clear invoices
As an affiliate, we earn on qualifying purchases.

TL;DR

OpenAI has announced a 50% price reduction for its GPT-6 Sol and Luna models, while their benchmark scores remain stable. This shift makes AI more affordable for a broad range of use cases, emphasizing cost efficiency without sacrificing performance.

OpenAI has announced a 50% reduction in the prices for its GPT-6 Sol and Luna models, effective immediately, while their performance benchmarks remain unchanged. This move aims to make high-quality AI more accessible for a broader range of applications, from enterprise workflows to consumer products.

The pricing for GPT-6 Sol now stands at $2.00 per 1 million input tokens (down from $4), and $10.00 per 1 million output tokens (down from $20). For GPT-6 Luna, prices are cut to $0.10 per 1 million input tokens (from $0.20) and $0.50 per 1 million output tokens (from $1.20). These reductions are attributed to improvements in caching and inference efficiency, which have lowered operational costs.

Despite the significant price cuts, independent evaluations, such as those from Artificial Analysis, confirm that the models’ performance scores have remained roughly stable. For example, GPT-6 Sol’s maximum effort score on the Artificial Analysis Intelligence Index is 48, well above the median of 25 for comparable models, with Luna scoring 37 against a median of 12. These models continue to demonstrate high cost-efficiency, with artificial metrics showing a halving of cost per task while performance metrics stay consistent or improve slightly.

OpenAI emphasizes that the reductions are aimed at democratizing access to advanced AI, especially for tasks where cost constraints previously limited deployment. The models’ improvements in caching, with 90% discounts on cached input reads, further support this shift by reducing ongoing operational costs.

At a glance
updateWhen: announced September 22, 2026
The developmentOpenAI has cut the prices of its GPT-6 Sol and Luna models by half, with no change in their benchmark scores, aiming to boost AI adoption and affordability.

GPT‑6 Sol and Luna: half the price, about the same intelligence

OpenAI’s September 22, 2026 release doesn’t raise the ceiling. It lowers the cost of everything below it, which changes what’s worth automating.

GPT‑6 Sol
$4 / $20 → $2 / $10
GPT‑6 Luna
$0.20 / $1.20 → $0.10 / $0.50

Per 1M input / output tokens. Cached input reads keep the 90% discount.

Cost per task, halved

Measured by Artificial Analysis as the weighted cost of one Intelligence Index task, at max effort.

GPT‑5.6 Sol
$1.99
GPT‑6 Sol
$1.06
GPT‑5.6 Luna
$0.18
GPT‑6 Luna
$0.07

The effort dial moves cost more than the model choice

Model and effortIntelligence IndexCost per task
GPT‑6 Sol (max)48$1.06
GPT‑6 Sol (low)34$0.13
GPT‑6 Luna (max)37$0.07
GPT‑6 Luna (low)21$0.0045
GPT‑6 Luna (non‑reasoning)18$0.01

Sol at low effort keeps about 70% of its max score for roughly an eighth of the cost, because it writes far fewer reasoning tokens. For reference, Claude Opus 5.5 leads the same index at 58.

What got better, and what got worse

Better

  • Hallucination rate on AA‑Omniscience: Sol 92% → 60%, Luna 93% → 77%
  • Coding Agent Index: Sol 57, up 2 points, at ~50% lower cost per task
  • OpenAI reports about half as many factual mistakes for Sol as its predecessor
  • Higher cache hit rates; GitHub reports over 50% fewer prompt tokens needing fresh processing

Sol gets there partly by declining more: it attempts 83% of questions vs 99%, and accuracy falls 59% → 54%.

Worse

  • GDPval‑AA v2.1: Sol down ~100 Elo, Luna down ~75
  • AA‑Briefcase v1.1: Luna down ~45 Elo
  • Coding Agent Index: Luna 41, down 2 points
  • Both models write more output tokens per task than their predecessors

Reviewers attribute the drops to weaker presentation and deliverables that omit required elements.

What to do about it

Already on GPT‑5.6 Sol or Luna? The move is mostly a price cut. Re‑test first if your output is a document someone reads, not data a system consumes.
Shelved an automation on cost? Token prices halved and the effort dial adds another order of magnitude. Re‑run the business case.
Choosing between labs? The question is no longer which model is smartest, but which clears your quality bar at the lowest cost per task.
ThorstenMeyerAI.comSources: OpenAI (pricing, vendor benchmarks) and Artificial Analysis (independent evaluation and model pages). Figures as of 23 September 2026.

Implications for AI Accessibility and Cost Management

The price reductions for GPT-6 Sol and Luna are significant because they lower the financial barrier for deploying large language models in various industries. For businesses, this means more affordable options for automating customer service, content generation, and research tasks, which previously might have been cost-prohibitive. The move aligns with OpenAI’s strategy to expand AI adoption by making high-quality models more economical, potentially accelerating innovation and integration across sectors.

Furthermore, maintaining performance benchmarks despite the price cuts demonstrates that cost efficiency does not necessarily come at the expense of quality. This could influence competitors to follow suit, leading to broader industry shifts towards more affordable AI solutions. For end-users, this means more accessible AI-powered tools with reliable performance, fostering wider adoption and new use cases.

Amazon

AI language model API subscription

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Background on OpenAI’s Pricing and Model Development

OpenAI’s release of GPT-6 Astra earlier this month marked a significant step in advancing AI capabilities, with Astra representing the top-tier, most capable model. The introduction of GPT-6 Sol and Luna on September 22, 2026, expanded the family by offering more cost-effective options aimed at scaling AI deployment. Prior to this, GPT-5.6 models were priced higher, limiting their use in budget-sensitive applications.

The company has consistently highlighted that improvements in caching, inference, and hardware utilization have enabled these price reductions. The move follows a broader industry trend of reducing costs to democratize AI access, with competitors like Anthropic also cutting prices recently. While performance levels have remained stable, some evaluations point to minor regressions in certain knowledge tasks, attributed to tuning for cost efficiency rather than raw accuracy.

OpenAI’s focus on balancing cost and performance reflects an ongoing effort to make AI tools more adaptable across different workflows, from enterprise to consumer markets. These developments come amid increasing demand for scalable, affordable AI solutions in a competitive landscape.

Amazon

cost-effective AI development tools

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Remaining Questions About Model Performance and Adoption

While benchmark scores remain stable, some evaluations, such as those on knowledge work tasks, have shown regressions, particularly in presentation quality and deliverables. It is unclear how these regressions will affect real-world applications that depend on detailed, polished outputs. Additionally, the long-term impact on market adoption and how competitors will respond remains uncertain, as does the potential for further price reductions or model improvements.

Amazon

AI inference caching solutions

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Next Steps for OpenAI and AI Market Response

OpenAI is expected to continue monitoring the models’ performance across various tasks and gather user feedback to refine tuning. The company may also introduce further updates to optimize cost-efficiency and address any identified regressions. Industry observers anticipate that competitors might respond with their own price cuts or feature enhancements. The broader AI market will likely see increased adoption of these more affordable models, especially in sectors where cost constraints previously limited AI deployment. OpenAI might also expand its caching and inference tools to further lower operational costs for developers and enterprises.

Amazon

enterprise AI model deployment

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Will the performance of GPT-6 Sol and Luna decline with the price cuts?

According to OpenAI and independent evaluations, benchmark scores have remained stable or improved slightly, though some knowledge tasks have seen minor regressions. Overall, performance appears consistent with prior models at the new price points.

How do the price reductions impact AI deployment in businesses?

The reductions significantly lower the cost per task, making it more feasible for companies to automate workflows, customer support, and content creation without exceeding budgets. This could accelerate AI adoption across industries.

Are there any trade-offs associated with the price cuts?

Some evaluations indicate regressions in presentation quality and detailed deliverables, suggesting that tuning for cost efficiency may slightly affect certain types of outputs. Users should test models for their specific needs before full deployment.

Will OpenAI introduce more models or further price cuts soon?

While no official announcements have been made, OpenAI’s focus on cost efficiency suggests ongoing efforts to optimize and possibly expand their model lineup or reduce prices further based on market response and technological improvements.

How do GPT-6 Sol and Luna compare to competitors like Anthropic’s Claude Opus 5.5?

Artificial Analysis reports that GPT-6 Sol and Luna offer competitive performance and significantly lower costs, with GPT-6 Sol scoring 48 and Luna 37 on the Intelligence Index, compared to Claude Opus 5.5’s top score of 58. Price-wise, OpenAI models are now more accessible, potentially offering better value for many use cases.

Source: ThorstenMeyerAI.com

FALL

Fall Picks

As an affiliate, we earn on qualifying purchases.

You May Also Like

12 AI-Powered Tools To Streamline Your Content Creation Process In 2026

Discover the top 12 AI-powered tools shaping content creation in 2026, streamlining research, drafting, and publishing for diverse channels.

The United States: The High-Variance Bet

The US is pursuing a minimal regulation, market-driven strategy for AI and social safety nets, with significant local experimentation amid federal inaction.

Is Canada The Future Of Europe’s AI Alliances?

European Commission’s proposal for closer ties with Canada could transform AI collaboration, integrating Canada’s advanced ecosystem into Europe’s strategic plans.

AI Surge Propels SenseTime To Its First Profitable Quarter With 617M Yuan

SenseTime reports its first-ever profit since going public, earning 617 million yuan due to rising demand for AI services in China, marking a key industry milestone.