News that matters

7 pages, most recent first

• Claude Haiku 5.5 at $0.10 per million tokens: which jobs an SMB can hand to it instead of a pricier model

  For short texts the saving is real. Above a certain prompt length the sums change, and there is a hidden price rise you should know about.

• Mistral Large 4: the European model you can already try via API, and when it makes sense to

  It's the first European model to come close to the best, but for now it's a preview and should be treated as one.

• Gemini 3.8 Flash TTS clones a voice in 30 seconds: what it can do for your business's phone line and videos

  For recorded messages and videos it's ready now and costs pennies. For talking to customers live on the phone, not yet.

• OpenAI dots: always-on agents, what access to give

  They work on their own across your apps even when nobody is watching. Before an employee switches one on, you decide what they can touch.

• Ollama: route requests with AI locally, at no cost

  Decision models don't write text: they pick a label and tell you how confident they are. For sorting emails and tickets, that is what you need.

• GPT-6.1 Sol costs a fifth of Astra: should you switch?

  The price list says yes for many business uses. OpenAI states the performance figures itself, so test them on your own cases before moving everything over.

• Claude Sonnet 5.5: same price list, but your bill depends on tokens

  The price per token hasn't changed. What you actually pay depends on how much the model writes, and you control that with the effort level.