Anthropic releases Claude Opus 5.5; OpenAI halves prices for GPT-6 Sol and Luna
Anthropic announced on September 22, 2026 that its new model Claude Opus 5.5 offers performance close to the top‑tier offering but at a lower cost. The company highlighted a new “Effort” setting that lets users choose how deeply the model thinks, and cited an example where the model rewrote legacy code with near‑top‑tier accuracy while saving time and money.
What you can do with it
AI at Work →Draft customer emails with GPT-6 Sol & Luna
What you get You can summarize meeting notes.
Handles structured tasks with clear instructions, reducing manual work.
- Who for
- marketers, support teams and operations
- Cost
- paid from $X (API pricing reduced by 50%)
- Effort
- minutes
Use it for
- Draft customer emails
- Summarize meeting notes
- Extract key points
How to set it up · From the article
- Set clear instructions (e.g., format, audience, character limit)
- Input documents or tasks via API
- Review AI output for accuracy before finalizing
Watch out Manual verification required for critical tasks like contracts or financial data
Key points from the news
- Anthropic's Claude Opus 5.5 promises near‑top‑tier performance with lower pricing and an "Effort" depth setting.
- OpenAI introduced GPT‑6 Sol and Luna, each priced at about half of earlier models.
- OpenAI reports reduced hallucinations but increased refusal to answer uncertain queries.
OpenAI revealed on the same day—reportedly only about 90 minutes later—that it is launching two GPT‑6 variants, Sol and Luna, each priced at roughly half of previous models. Sol is aimed at tasks requiring deep reasoning such as writing or coding, while Luna targets high‑volume, lighter tasks like document summarisation. OpenAI also said hallucination rates have dropped, though the models now more often refuse to answer when uncertain.
Both announcements stress affordability over raw intelligence, a shift that could make generative AI more accessible for everyday work and personal use. The simultaneous price cuts suggest a competitive push to broaden adoption among users who were previously deterred by cost.
Model pages: Claude Opus 5.5 → · GPT-6 Sol → · GPT-6 Luna →
The story so far
2 episodes →- Anthropic releases Claude Opus 5.5; OpenAI halves prices for GPT-6 Sol and Lunathis story
Gentle AI Newspaper - For those in their 50s and 60s / Easy for beginners
note.com · 24 September 2026
Loading the full article…
This text was published by note.com and written by K-Naoki ペットとAIを愛する男. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
11sources- OpenAI's GPT-6 Just Got 50% Cheaper — Box CEO Says AI Agent Opportunity Could 'Dramatically' Expand: 'What an Insane Day'Press · finance.yahoo.com ·
- How to choose between GPT-6 Sol/Luna and Claude Opus 5.5? — AI model selection in autumn 2026 shifts from 'the strongest' to 'division of labor'Press · note.com ·
- [Breaking News] New ChatGPT and Claude models are insane! The concept of "doing work" with AI is changingPress · note.com ·
- GPT-6 Sol and Luna | New ChatGPT models, API pricing is half that of the previous generationPress · note.com ·
- GPT-6 Sol and Luna Finally Announced. What Changes for Corporate AI Adoption?Press · note.com ·
- OpenAI Launches GPT-6 Sol and Luna With 50% Lower API CostsPress · propakistani.pk ·
- GPT-6 Sol and Luna: Which One Should You Choose? — Three Levels of 'Work Weight' Including AstraPress · note.com ·
- AI is shifting from a 'tool to ask questions' to a 'tool to put to work repeatedly.' What the GPT-6 pricing revealsPress · note.com ·
- What Changes with GPT-6 Sol & Luna? AI Utilization for SMEs and Kyushu AI Management | AI Ryohin, Fukuoka FDEPress · note.com ·
- OpenAI Introduces GPT-6 Sol And Luna, Slashing API Prices 50% To Bring Advanced AI To More UsersPress · finance.yahoo.com ·
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- OpenAI releases ChatGPT Images 2.5 with sketch, point, and template tools · 4 src
- FreedomIntelligence releases HuatuoGPT-3 multimodal model on Hugging Face · 1 src
- PrismML launches 1-bit Bonsai LLM for Qualcomm smart glasses · 1 src
- BottleCap AI cuts Qwen3.8-27B’s reasoning tokens by 37.2% with ThinkingCap-Qwen3.8-27B · 1 src
- What has changed with GPT-6 Astra? Basic specifications of the new model that supports 'thinking, researching, and creating' in ChatGPT · 2 src
Comments
via GitHub Discussions