Skip to content
All news
Release1 min read

New models: GPT-5.6, Claude Opus 5, Gemini 3.7 Flash and Grok 4.6

The model line-up moved to the current versions from OpenAI, Anthropic, Google and xAI, with Flux 2 Pro and Stable Diffusion 3.5 Large for images.

Milly Lab now serves the current versions of the models you use most. Your chats, plans and settings stay as they were: the same model choice simply points to the newer version.

Chat models #

  • GPT-5.6 Luna, Terra and Sol from OpenAI: a fast model, an everyday model and the flagship.
  • Claude Opus 5 from Anthropic, next to Claude Sonnet 5 and Claude Haiku 4.5.
  • Gemini 3.7 Flash from Google, also in a version with a 200K-token context.
  • Grok 4.6 from xAI.
  • DeepSeek V4, V4 Flash and V4 Pro stay in the line-up.

Image models #

Light Studio moved to Flux 2 Pro, Stable Diffusion 3.5 Large and Gemini Nano Banana 2.

Which models your plan includes #

Each plan includes exactly the models it lists, each with its own monthly allowance. The Models page shows every model and what it is good at.

Share

More news

Release

Workflows are on for every account

Draw a repeatable process once (a trigger, the steps and the approvals it needs) and let Milly Lab run it on a schedule, from a form or from a chat. Fifteen ready-made templates help you start.

Release

A new sign-in, with codes by e-mail

Signing in and creating an account now happen on one redesigned screen, and two-factor sign-in can use a 6-digit code sent to your e-mail.

Publication

Milly Lab has a new home at m-lab.app

Our company site now lives at m-lab.app in English, Russian and Tajik, and the workspace itself has its own address, app.m-lab.app.

See what is new for yourself

Every account starts on the Free plan, without a card.