Skip to content

A practical perspective for individuals

Less monitoring. More value.

We turn new tools and features into one question: what can you do with them today, without unnecessary hype?

  • AvailabilityIs it actually out?
  • BenefitWhat work could this improve?
  • First stepWhat is worth verifying?
938 practical events in the selection 72 active sources Updated

Practical radar

What deserves your attention

Browse the archive →
GitHub important Productivity

GitHub made the Customize tab in the GitHub Copilot app available to all users

The Customize tab in the GitHub Copilot app is now generally available. It brings MCP servers, plugins, skills and canvases together in one place, offering both a recommended Featured selection and browsing by type.

What you get out of it

Users of the GitHub Copilot app can find MCP servers, plugins, skills and canvases in one place and more easily adapt the tool to their own way of working, without having to know in advance which specific customization they need.

✓ official
Details and sources →
important New models

RadixArk releases open-weight model DaisyRose-27B for agentic tasks

RadixArk has released the DaisyRose-27B model under the Apache 2.0 license, a fine-tuned version of the Qwen3.8-27B model focused on long agentic tasks. On the Terminal Bench 4.1 benchmark it improved pass@1 from 5.6% to 12.0%, and other coding and instruction-following tests also showed improvement.

What you get out of it

Developers and people experimenting with AI agents for programming are getting a new open model available for download and local deployment, with an architecture and 256K-token context window identical to the Qwen3.8-27B model.

✓ official
Details and sources →
OpenAI important New models

GitHub Copilot rolls out OpenAI's GPT-6.2 Astra and GPT-6.2 Sol models

GitHub Copilot now offers OpenAI's GPT-6.2 Astra and GPT-6.2 Sol models for agentic coding and code review. According to GitHub, in tests they caught more issues than the older GPT-6 models while consuming fewer tokens. Available for Pro+, Max, Business, and Enterprise, with a gradual rollout.

What you get out of it

Developers with a Copilot Pro+ or Max plan can try the GPT-6.2 Astra and GPT-6.2 Sol models in the model picker for code review and agentic tasks, which according to GitHub offer better issue detection and lower token consumption than the older GPT-6 models.

✓ official
Details and sources →
important New models

Experimental Granite 4.1 models with the new LLMON markup language against prompt injection attacks

The preview models granite-4.1-3b-LLMON-preview and granite-4.1-8b-LLMON-preview have been released, trained on the new LLMON format, which explicitly separates instructions from data and, according to the source, reduces the success rate of indirect prompt injection attacks to nearly zero.

What you get out of it

Developers working with Granite models can try the preview models and the accompanying toolkit (including a JSON→LLMON converter) to reduce the risk of prompt injection in their own LLM applications.

✓ official
Details and sources →
important New models

Liquid AI releases weights for d1-3B and d1-omni-600M for local decision-making

Liquid AI has released the open weights of d1-3B and d1-omni-600M. The models return decisions in a single pass without generating tokens. They support text with images, and the experimental d1-omni-600M also supports text with audio.

What you get out of it

Developers can try decision-making based on text and images on their own devices; d1-omni-600M is available as an early research version for experiments with text and audio.

✓ official
Details and sources →
OpenAI important Tools and apps

OpenAI enabled setting expiration dates for API keys and enforcing their maximum lifetime

OpenAI added the option to set an expiration period for API keys when creating them in OpenAI API. Administrators can also enforce a maximum lifetime at the organization or project level in Platform settings.

What you get out of it

Developers working with OpenAI API can set an expiration date when creating a new key, reducing the risk posed by a forgotten key that remains valid for a long time.

✓ official
Details and sources →
important New models

IBM has released the open-source model Granite Time Series PatchTST-FM-r2 for time series forecasting

IBM has released the open-source model Granite Time Series PatchTST-FM-r2 (~385M parameters) for zero-shot time series forecasting. According to the company, it is the best permissively licensed model on the GIFT-Eval leaderboard, available under Apache 2.0/OpenMDW 1.0 for commercial use.

What you get out of it

Data scientists and ML engineers working on time series forecasting gain a freely available model that can be deployed zero-shot instead of training a separate model for each dataset.

✓ official
Details and sources →
GitHub important AI agents

GitHub Copilot Code Review can now approve pull requests automatically

GitHub has launched a feature in public preview that allows Copilot Code Review to approve pull requests automatically. It is disabled by default and can be enabled at the enterprise, organization, or repository level.

What you get out of it

Developers working with pull requests on GitHub will now also see an assessment of whether the pull request is ready for approval in reviews from Copilot Code Review, and if an administrator enables the feature, this approval can replace part of the human review.

✓ official
Details and sources →
important update New models

Apertus 1.5 models add thinking mode; speech processing is experimental

The Apertus 1.5 models have a switchable thinking mode; speech processing remains experimental. The 8B variant underwent further training on 4 trillion tokens, and the 70B variant on 2 trillion. The totals reached 19 and 17 trillion tokens.

New: The 8B variant adds 4 trillion training tokens for a total of 19 trillion.; The 70B variant adds 2 trillion training tokens for a total of 17 trillion.; Spoken language processing is experimental.; The models can be switched into thinking mode.; The models were trained for better integration of external tools and APIs.

What you get out of it

When working with recordings, you can obtain a text response, but the experimental status of speech processing is a reason to check its accuracy against the original recording.

✓ official
Details and sources →
important New models

Liquid AI releases LFM2.5-24B-A2B with open weights for local agents

Liquid AI has made LFM2.5-24B-A2B available for local agents. It has 24 billion parameters, with 2 billion active per token, and a context window of 131 072 tokens. According to the company, it generates 184 tokens per second on Apple M5 Max hardware.

What you get out of it

An individual can try a local agent to work with their own material and take advantage of the larger context window compared with the previous model.

✓ official
Details and sources →
Anthropic important update Security

OSS Scanner is intended to suggest fixes, and Anthropic expects accuracy above 90 %

The free OSS Scanner tool is intended to suggest fixes alongside detecting vulnerabilities. Anthropic expects accuracy above 90 %, but the reports do not undergo human review. Maintainers of open-source projects critical to infrastructure or user safety can sign up through GitHub.

Part of an overview of multiple AI topics; only this event has been covered.

New: The scanner is intended to suggest fixes for vulnerabilities.; Anthropic expects scanner accuracy above 90 %.; Maintainers can sign up through GitHub.; Enrollment is intended for projects critical to infrastructure or user safety.

What you get out of it

A maintainer of an open-source project critical to infrastructure or user safety may receive regular checks and suggested fixes without a service fee.

✓ official
Details and sources →
Google important AI agents

Google has made a work agent available in the Gemini Enterprise app in private preview

Google has made a work agent available to enterprise customers in the Gemini Enterprise app in private preview. According to the company, it can plan tasks, work across connected systems and act under its own identity with an audit trail.

What you get out of it

Employees with access to the private preview can assign work tasks from a single interface and track their progress in the “tasks inbox” interface. According to Google, context is also preserved when switching between connected devices.

✓ 2
Details and sources →

When personal experience proves useful

Does it make sense for a team too?

Look at the same topics through the lens of processes, risks, and implementation within a company.

Go to the company radar →