AIThis post was created with the assistance of artificial intelligence (AI).

🔍 Read the full analysis: OpenAI’s Price Drop For GPT‑6 Sol And Luna: What’s The Same And What’s Changed? 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, focusing on making AI more affordable for businesses and developers. The models maintain comparable performance levels, with some quality trade-offs, marking a shift toward cost-effective AI deployment.

OpenAI has officially slashed the prices of its GPT‑6 Sol and Luna models by 50%, marking a significant shift toward more accessible AI solutions for businesses and developers. The models, introduced on September 22, 2026, are now priced at half of their GPT‑5.6 predecessors, with the company emphasizing cost efficiency as a key driver behind this move.

The new pricing reflects improvements in caching and inference technology, enabling OpenAI to serve these models at lower costs, which are then passed on to users. GPT‑6 Sol now costs $2.00 per 1 million input tokens (down from $4), and $10.00 per 1 million output tokens (down from $20). GPT‑6 Luna is priced at $0.10 per 1 million input tokens (from $0.20) and $0.50 per 1 million output tokens (from $1.20). These reductions are roughly 50% compared to GPT‑5.6 promotional pricing.

Independent analysis from Artificial Analysis confirms that while costs have halved, model performance remains roughly stable, with some metrics showing slight regressions. The models demonstrate improved hallucination mitigation, with Sol reducing hallucination rates from 92% to 60%, and Luna from 93% to 77%. However, some evaluations indicate regressions in knowledge work tasks, attributed to changes in output presentation quality.

At a glance
updateWhen: announced September 22, 2026
The developmentOpenAI released GPT‑6 Sol and Luna on September 22, 2026, with prices halved from previous GPT‑5.6 models, aiming to improve affordability without major performance drops.

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.

Impact of Price Reduction on AI Deployment

The 50% price cut makes advanced AI models more accessible for a broader range of applications, especially for workflows where cost constraints previously limited adoption. This shift could accelerate AI integration across industries, enabling smaller firms or projects with limited budgets to leverage high-performance language models. However, the performance trade-offs in some tasks highlight the importance of careful evaluation before deployment in critical workflows.

Amazon

AI model deployment tools

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Background on GPT‑6 Model Pricing and Capabilities

OpenAI’s GPT‑6 models, introduced as part of the Astra family, marked a significant advancement in AI capabilities, with larger context windows and improved inference efficiency. Prior to this update, GPT‑6 models were priced higher, reflecting their enhanced performance. The recent price reductions aim to democratize access, aligning with OpenAI’s broader strategy to distribute AI benefits more widely. The models’ performance remains competitive, with some evaluations showing stability and others noting minor regressions in specific tasks, particularly those requiring detailed output presentation.

Amazon

cost-effective AI development software

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Unresolved Aspects of Model Performance and Adoption

While cost reductions are confirmed, the long-term impact on model quality across diverse use cases remains uncertain. Some evaluations indicate regressions in detailed knowledge tasks, raising questions about the models’ suitability for certain workflows. Additionally, the extent to which these price cuts will influence broader market adoption and competitive responses is still developing.

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 Users

OpenAI is expected to continue refining its models, potentially balancing cost and quality further. Users should monitor upcoming updates, especially regarding model tuning for specific tasks. Industry observers anticipate increased adoption of GPT‑6 models in commercial applications, with ongoing evaluations to assess real-world performance and cost-effectiveness. OpenAI might also expand caching and inference optimizations to further lower operational costs.

Amazon

large language model API access

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

How much cheaper are GPT‑6 Sol and Luna now?

GPT‑6 Sol is priced at $2.00 per 1 million input tokens and $10.00 per 1 million output tokens, while GPT‑6 Luna costs $0.10 and $0.50 per million tokens, respectively. These are approximately 50% less than previous GPT‑5.6 prices.

Do performance levels remain the same after the price cut?

Independent analysis suggests that while costs have halved, performance remains roughly stable overall, with some improvements in hallucination reduction but minor regressions in knowledge work tasks.

What are the main trade-offs with these models?

The models show a trade-off between hallucination rates and answer completeness, with Sol attempting fewer questions to reduce errors, which can lead to less detailed outputs in some cases.

Will this price reduction affect AI adoption in industry?

Yes, the lower prices could significantly increase adoption, especially for smaller organizations and use cases previously limited by cost, although quality considerations remain relevant for critical applications.

What improvements have been made technically to enable lower prices?

OpenAI credits advancements in caching and inference technology, which allow more efficient reuse of context and lower operational costs, enabling the price reductions.

Source: ThorstenMeyerAI.com

FALL

Fall Picks

As an affiliate, we earn on qualifying purchases.

You May Also Like

Asahi Linux On M3

Asahi Linux is making strides in supporting Apple M3 chips, with recent developments indicating early compatibility efforts, though full support remains unconfirmed.

Mac vs GPU Tower for Local LLMs: The Heat-and-Noise Tradeoff

Comparing Mac Studio with Apple Silicon and GPU towers for local large language model inference reveals key heat, noise, and capacity differences affecting choice.

The Role of AI in Optimizing Electric Bus Routes

Journey into how AI transforms electric bus routing, unlocking smarter, more efficient transit—discover the innovations that could change transportation forever.

Wireless Inductive Charging: Advances and Real‑World Deployments

Optimized for convenience and efficiency, wireless inductive charging advancements are transforming everyday device use—discover how these innovations are shaping our world.