Google Gemma 3 Med 27B
Large medical variant for complex clinical reasoning and biomedical analysis.
Gemma 3 Med 27B is an open-weight text model built by Google. It is multimodal: alongside a text prompt it accepts images, then replies with generated text. It runs entirely on your own hardware through Ollama: no API key and no per-use fees.
- Released
- Jul 2025
- Pricing
- Free, runs on your hardware
- Inputs
- Text, Images
Samples
Generated with Gemma 3 Med 27B via Ollama. The same three prompts run against every text model in the catalog, shown verbatim, so the only thing that changes between two models’ answers is the model.
Explain what an API is to someone who has never written code. Do not use a restaurant or waiter analogy. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 50 words.
An API (Application Programming Interface) is like a digital messenger. It allows different software programs or systems to talk to each other and exchange information without needing to know the complex details of how the other works internally. It simply requests and receives specific data or actions.
A two-person startup is deciding whether to build their own user authentication or use a hosted provider. Give them a clear recommendation, then the single strongest argument against your own recommendation. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 150 words.
Use a hosted provider. For a two-person startup, building and maintaining secure user authentication is a huge time sink and security risk. Hosted providers like Auth0, Firebase Authentication, or Okta handle the complexities of security, scalability, and compliance, allowing you to focus on your core product. You can integrate them quickly and easily. The strongest argument against this is vendor lock-in. Relying on a third party means you're dependent on their pricing, availability, and roadmap. If you later decide to switch, migrating users can be complex and potentially disruptive. However, the immediate benefits of speed, security, and reduced operational overhead typically outweigh this risk for an early-stage startup.
Write a briefing for a non-technical executive explaining why their company's spending on AI is hard to predict, and what they can do about it. Cover at least three distinct causes. Plain prose only — no headings, bullet points, or markdown formatting. Write exactly 205 words.
Good morning. We need to discuss the unpredictability of our AI spending. It's proving difficult to forecast accurately, and I want to explain why. First, AI is a rapidly evolving field. New tools, techniques, and platforms emerge constantly. We might invest in a solution today, only to find a superior or more cost-effective alternative available next quarter. This makes long-term budget planning challenging. Second, the scope of AI projects is often unclear initially. We might start with a well-defined goal, but as we explore the data and capabilities, the project scope can expand significantly, leading to unexpected resource needs. This is especially true when dealing with complex business problems where the full potential of AI isn't immediately apparent. Third, AI talent is in high demand and short supply. The cost of hiring skilled data scientists, engineers, and researchers fluctuates significantly based on market conditions. Retaining this talent also requires competitive compensation and benefits, which can be unpredictable. To manage this better, we should adopt a more agile approach. Start with smaller, proof-of-concept projects to better understand costs and benefits before large-scale investments. Invest in internal training to reduce reliance on expensive external consultants. Finally, build flexibility into the budget to accommodate unforeseen opportunities or challenges in this dynamic field. This will allow us to adapt and maximize our AI investments effectively.
Google and Google DeepMind build the Gemini family of multimodal models, the Imagen and Nano Banana image models, the Lyria music models, and the Veo video models.
deepmind.google ↗Google's flagship reasoning model for complex software engineering, deep mathematical and scientific problem solving and autonomous multi-step agents, with a shared 1M-token context.
Google's most capable Flash model, with clear gains over 3.7 Flash in software engineering, agentic tasks, and multi-step reasoning, plus vision and a 1M-token context window.
Google's most efficient multimodal Flash model: an updated reasoning stack with stronger coding and computer-use quality, at the same speed and scale as the rest of the Flash line.
Google's most capable multimodal model with deep reasoning and 1M token context.
Google's 26B mixture-of-experts model with long context and strong multimodal reasoning.
Google's flagship 31B dense model with 256K context and top-tier text and image understanding.