Anthropic and OpenAI roll out cheaper models in first release since call for slowdown
- ID
- 27375
- Status
- summarized
- Published
- 23 Sep 2026, 3:43 AM
- Fetched
- 23 Sep 2026, 4:03 AM
- Provider
- CNBC Technology
- Category
- technology
- Original URL
- https://www.cnbc.com/2026/09/22/anthropic-openai-cheaper-ai-models.html
- Source URL
- https://www.cnbc.com/id/19854910/device/rss/rss.html
Summary
- Score
- 8.0
- Created
- 23 Sep 2026, 4:04 AM
- Tags
- Audience
- developersvibe_codersai_ml_learnersai_agent_userssaas_founders
What happened
OpenAI launched GPT-6 Sol (for coding workloads, tier below Astra) and GPT-6 Luna (for high-volume tasks like extraction and summarization), cutting API prices 50% versus GPT-5.6 promotional pricing. Anthropic released Claude Opus 5.5, a more token-efficient model costing roughly 40% less than Opus 5. Both moves come amid pressure from cheaper open-weight competitors.
Why it matters
If you're paying for frontier model API calls in production, re-run your cost projections now—50% and 40% drops are large enough to change which tier you route workloads to. Sol being positioned specifically for coding means you should benchmark it against your current coding model before defaulting to the top tier. For Malaysian SaaS founders shipping AI features, this materially lowers the per-request cost floor and may make previously uneconomical agent workflows viable.
Discussion angle
Compare Sol vs Claude Opus 5.5 for coding workloads at the new price points—does the 50% discount on Sol make it the default for agent-based coding pipelines, or does Opus 5.5's token efficiency win on total cost for longer contexts?