Google DeepMind ships Gemini 3.8 with real-time and extended reasoning modes
Google DeepMind has released Gemini 3.8 Live and 3.8 Live Extended Thinking, signaling continued iteration on real-time conversational AI and reasoning capabilities. The Live variant targets low-latency interaction, while Extended Thinking deepens the model's ability to work through complex problems before responding. This dual-track release reflects the industry's push toward both speed and depth in reasoning, positioning Gemini against competitors investing heavily in inference-time compute. For practitioners, the availability of extended reasoning in a production model expands use cases in technical domains where deliberation matters more than immediate response.
Modelwire context
Skeptical readThe announcement bundles two meaningfully different capabilities under a single release, which risks obscuring whether either has been independently validated. Notably, Google does not specify what 'extended thinking' actually costs in latency or tokens, a detail that matters enormously for practitioners deciding whether to use it in production.
This release has no direct anchor in our current archive, so it sits in a broader pattern worth naming: the major labs have converged on a two-track product strategy, offering fast inference for conversational use and slower deliberative reasoning for complex tasks. That pattern has been visible across OpenAI and Anthropic releases over the past year, and Google is now explicitly mirroring it. Whether Gemini 3.8 Live's latency profile is genuinely competitive, or whether Extended Thinking's reasoning gains hold outside curated demos, remains unverified from this announcement alone.
Watch for third-party evaluations on GPQA Diamond or MATH-500 within the next 60 days. If Extended Thinking's gains replicate on held-out splits that weren't available before the release date, the capability claim is credible. If they don't appear, the benchmarks in the announcement deserve closer scrutiny.
This analysis is generated by Modelwire’s editorial layer from our archive and the summary above. It is not a substitute for the original reporting. How we write it.
MentionsGoogle DeepMind · Gemini 3.8 · Gemini 3.8 Live · Gemini 3.8 Live Extended Thinking
Modelwire Editorial
This synthesis and analysis was prepared by the Modelwire editorial team. We use advanced language models to read, ground, and connect the day’s most significant AI developments, providing original strategic context that helps practitioners and leaders stay ahead of the frontier.
Modelwire summarizes, we don’t republish. Google DeepMind originally reported this story as “Introducing Gemini 3.8 Live and 3.8 Live Extended Thinking”. The full content lives on deepmind.google. If you’re a publisher and want a different summarization policy for your work, see our takedown page.