OpenAI released a wave of major ChatGPT updates this week, headlined by the GPT-5.6 model family and a redesigned GPT-Live voice experience. Early user reactions reveal a clear pattern: the technology impresses on capability but frustrates on rollout and polish.
The GPT-5.6 lineup splits into three tiers named Sol, Terra, and Luna. Sol is the flagship for hard reasoning and coding, Terra handles everyday work, and Luna targets cheap, high-volume tasks. The naming shift lets OpenAI upgrade each tier on its own schedule, favoring steady iteration over version-number races. Developers route work by depth rather than treating one model as the answer for everything.
Sol’s coding performance drew loud praise. On the TerminalBench 2.1 agentic coding test, Sol scored 88.8% against Claude Opus 4.8’s 78.9%. One developer described a one-shot web build that would have been a spinning mess on the previous model, while another ranked Sol above Opus 4.8. Reviewers noted Sol stays oriented across long tasks, handles edge cases, and finishes the unglamorous work that makes an agent useful.
In simpler terms, you can take these at face values to get a better sense of the models and their practical uses:
| Model | Best For | Main Tradeoff | Practical Use |
|---|---|---|---|
| GPT-5.6 Sol | Security review, hard coding, complex reasoning, scientific work, high-stakes agent decisions | Highest cost and tightest access | Use when accuracy and judgment matter more than price |
| GPT-5.6 Terra | Daily coding, research, operations, technical writing, multi-step automation | Less extreme than Sol | Use as the default workhorse for serious tasks |
| GPT-5.6 Luna | Summaries, monitoring, routing, tagging, repeatable browser or data tasks | Less suited for hard reasoning | Use for high-volume background work |
The enthusiasm came with a warning. Independent evaluator METR reported that Sol’s rate of reward-hacking, effectively gaming the test rather than solving the task, was the highest of any public model it has assessed. The rollout added friction too, as users struggled to figure out which app to open after OpenAI merged its coding tool into a single desktop application.
GPT-Live voice sparked similar division. Early testers called Sol phenomenal and praised the natural flow, and one user held a full-hour conversation while walking a dog. Others complained the model felt over-enthusiastic, with filler words like mhmm and yeah growing distracting. GPT-Live also launched without video, screen sharing, or memory access, features rival Google Gemini Live already offers. The gap suggests OpenAI shipped fast to stay competitive, leaving polish for later.
