OpenAI releases GPT-5.6 in three versions as COO Fidji Simo departs after 11 months
OpenAI released GPT-5.6 Thursday in three versions—Luna, Terra, and Sol—with Sol claiming benchmark wins over Anthropic's Claude Fable on coding tasks. The launch coincides with COO Fidji Simo's departure less than a year after joining, citing worsening health issues.
OpenAI Ships GPT-5.6 in Three Tiers
OpenAI released GPT-5.6 on Thursday in three versions: Luna, Terra, and Sol, corresponding to increasing capability levels. According to OpenAI's blog post, Sol outperformed Anthropic's Claude Fable on the Agents' Last Exam and Artificial Analysis Coding Agent Index benchmarks. Specific scores were not disclosed.
The release follows earlier announcements this week including a new voice model called GPT-Live, integration of Codex into the desktop app, introduction of the ChatGPT Work agent, and retirement of the Atlas browser.
Early users reported speed improvements and enhanced reasoning capabilities. Box CEO Aaron Levie described Sol as "a big step up from GPT-5.5, especially on complex data-oriented tasks that require deep reasoning and analysis."
Katie Parrott from Every called GPT-5.6 "our favorite model to collaborate with," though noted "Fable still gets the assignments we want to hand off completely." She characterized the model as "fast enough to keep up with you, resourceful enough to find the context it needs to do good work, persistent when the first approach fails."
Sol vs. Fable: Split Verdict
User opinions on how GPT-5.6 Sol compares to Claude Fable varied widely. Ghostty creator Mitchell Hashimoto expressed a general preference for Sol, citing faster performance and better overall work output, while still reaching for Fable for "highly targeted debug or performance work with clear reward functions."
Figma CEO Dylan Field discouraged direct comparisons, calling them "apples and oranges," though he added Sol is "very good."
Wharton professor Ethan Mollick concluded that "both Sol & Fable represent jumps over previous models and have opened a large gap with the next-best AIs... if you [are] doing any work where better intelligence matters, those two models are your only choices."
Pricing for the three GPT-5.6 tiers was not disclosed.
Executive Departure Continues Pattern
COO Fidji Simo informed staff Thursday she will step down less than a year after joining OpenAI as CEO of applications, the Wall Street Journal reported. Simo cited worsening symptoms of postural orthostatic tachycardia syndrome, a chronic illness. She will become an adviser to the company.
Simo joins multiple high-profile departures from OpenAI in recent months, including chief futurist Joshua Achiam, VP of research Jerry Tworek, former chief product officer Kevin Weil, head of enterprise sales Barret Zoph, model behavior lead Joanne Jang, and researcher Max Schwarzer.
Meta's Muse Spark 1.1
Meta also released Muse Spark 1.1 on Thursday, though details about capabilities, benchmarks, and pricing were not provided in available reporting.
What This Means
The simultaneous release of major models from OpenAI and Meta marks an intensifying competition at the frontier of AI capabilities. Early user feedback suggests GPT-5.6 Sol and Claude Fable have established a two-model lead over competitors, though use cases may vary between them. The continued executive churn at OpenAI—now including its No. 2 leader—raises questions about organizational stability even as the company ships competitive products.
Related Articles
OpenAI releases GPT-5.6 with three model variants, claims 80-point Coding Agent Index score for Sol
OpenAI released GPT-5.6 in three variants: Sol ($5 input/$30 output per 1M tokens), Terra ($2.50/$15), and Luna ($1/$6). According to OpenAI, Sol achieves an 80-point score on the Artificial Analysis Coding Agent Index, 2.8 points above Anthropic's Fable 5, while using less than half the output tokens and costing one-third less.
OpenAI announces GPT-5.6 with three models (Sol, Terra, Luna) and ChatGPT Work agent tool
OpenAI released GPT-5.6 in three model tiers—Sol (flagship reasoning), Terra (mainstream), and Luna (instant)—positioning them against Anthropic's Claude models. The company claims GPT-5.6 Sol scores 53.6 on Agents' Last Exam, 13.1 points above Claude Fable 5, while completing tasks 61% faster. ChatGPT Work, a desktop productivity agent similar to Claude Cowork, launches simultaneously for Pro, Enterprise, and Edu users.
OpenAI releases GPT-5.6 family in three sizes: Luna at $1/$6, Terra at $2.50/$15, Sol at $5/$30 per 1M tokens
OpenAI released its GPT-5.6 flagship model family in three sizes: Luna ($1/$6 per 1M tokens), Terra ($2.50/$15), and Sol ($5/$30). The company claims GPT-5.6 Sol scores 53.6 on the Agents' Last Exam benchmark, outperforming Claude Fable 5's score by 13.1 points.
OpenAI designates GPT-5.6 as 'preferred model' for Microsoft 365 Copilot following cost-cutting reports
OpenAI has designated its newly launched GPT-5.6 as the 'preferred model' for Microsoft 365 Copilot, powering Word, Excel, PowerPoint, and Cowork. The announcement comes days after Bloomberg reported Microsoft is increasingly deploying its in-house MAI models to reduce reliance on OpenAI and cut costs.
Comments
Loading...