Google DeepMind launches Gemini 3.6 Flash, confirms Gemini 4
Google DeepMind has launched three new Gemini Flash models, including Gemini 3.6 Flash with enhanced performance and lower costs. Simultaneously, the company confirmed the development of an entirely new foundation model, Gemini 4.
What happened?
On Tuesday, 22 July 2026, Google DeepMind launched three new Gemini Flash models: Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, and Gemini 3.5 Flash Cyber. Gemini 3.6 Flash offers improved coding capabilities, knowledge management, and multimodal performance. The company also announced it has commenced its most ambitious pre-training programme to date for a new model named Gemini 4.
Key facts
”Google DeepMind on Tuesday released three new Gemini Flash models — Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, and a restricted cybersecurity specialist called Gemini 3.5 Flash Cyber — and buried inside the same announcement was a disclosure that may matter more than any of them: t”
”Beginning a new pretraining run — the most compute-intensive and irreversible phase of building a large language model — means the company decided the problem was architectural, not fixable through fine-tuning or additional data curation.”
Why it matters
The launch of the new Flash models, combined with the confirmation of Gemini 4, indicates a strategic reorientation. This comes against the backdrop of reports that the previously anticipated Gemini 3.5 Pro model failed to meet internal expectations. The decision to initiate new pre-training suggests that Google assesses that the existing architecture cannot be further optimised to match competitors.
Who is affected?
Developers utilising AI models can benefit from the new Flash models' improved efficiency and lower token costs. Companies invested in or dependent on Google's AI ecosystem are also directly affected. Users interacting with AI-driven applications can expect enhanced experiences as the new models are implemented.
What else you should know
Bloomberg reported on 16 July 2026 regarding a delay of Gemini 3.5 Pro and its failure to meet internal coding benchmarks, despite an update to training data in late June.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Vilka modeller har lanserats?
The link opens in a new window and leads to the publisher's own site.
Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.
AI-verktyg i artikeln
Topics
Get similar news straight to your inbox
The reader's room
Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.
Sign in to submit a comment or question.
Read the article through your role
- Assess technical risk: model choice, vendor lock-in, data flow and running cost.
- Update the architecture doc if new APIs or regulations touch production.
- Ensure observability + rollback plan before rolling out to production.
Generated angle — not editorial analysis of "Google DeepMind launches Gemini 3.6 Flash, confirms Gemini 4"