SupaKeeper Plain comparisons of AI coding tools

How to find out when to try new models for AI hobbyists in 2026

Learn how AI hobbyists can read estimated release odds, check how those estimates are built, and plan a test before a new model arrives.

By SupaKeeper ยท

SupaKeeper is published by LunarWerx, which makes some of the tools compared here.

To find out when to try new models, follow an estimated chance of release, learn how that estimate is built, and have a test ready before a launch. Model Odds is one option that ranks Anthropic and OpenAI model families by estimated chance of release within 30 days. [1]

OptionWhat it can doWhat to check
Model OddsRanks Anthropic and OpenAI model families by estimated chance of a release within 30 days. [1]Check whether a 30 day window fits how far ahead you plan to test. Read how the estimate is built before acting on it.
Reading company posts yourselfLets you read each company's own posts directly.Check how often you will read them and whether you would notice a release in time.
General news searchLets you search for recent news about a model name.Check that results carry dates and come from sources you trust.

What you need before you start

Before you look for news of a new model, know which companies and model families you care about. A hobbyist who uses one provider needs a different view than someone comparing several.

Have a short list of the models you already use and the tasks you want to try next. That makes it easier to decide whether a new release is worth your time when it appears.

Decide how much notice you need. Some people want to hear about a release early enough to read about it before trying it, while others only want to know when to check back. If you want alerts, have an email address ready that you can confirm.

Finally, plan a small test you can run in an afternoon, such as the same prompt on the old model and the new one. A test that is ready in advance means a release does not catch you without a plan.

Step by step

  1. Write down the model families you use or want to try, and the company that makes each one.
  1. Look for a source that gives an estimated chance of release, not only a list of past news. Compare how often that source updates with how quickly you expect to act on what it says.
  1. Check how the estimate is built. Ask whether it uses past launch dates, whether each past release is confirmed by a source you can read, and whether it shows how much its estimates disagree.
  1. Set a rough threshold for action, such as a chance you find high enough to justify a test, and write it down before a release happens.
  1. When a model crosses your threshold, run your prepared test on it and note the results beside the same test on your current model.
  1. Keep a record of past estimates against what actually happened. Over time you will see which sources you trust for timing.
  1. Recheck your list when the models you use change, so your watch stays matched to your hobby.

Tools that can do it

Model Odds is made by the team that publishes this site.

Model Odds ranks every Anthropic and OpenAI model family by its estimated chance of a release within 30 days. [1] Anthropic and OpenAI each get their own page that lists every past release. [1] Each past launch the model uses has to be confirmed by an official post and an independent timeline, and it is listed with its date. [1]

The estimate is built from the gaps between recent launches, and it also takes into account how long the current wait has already lasted. [1] It runs three estimators side by side and shows the spread between them as an error bar, a visible range of uncertainty. [1] Signals read automatically from company developer kits and docs, and from third-party model catalogues when two of them agree, can only raise the odds. [1]

The estimator can replay any past date without using data from after that date, so you can see how it would have judged earlier releases. [1] You can sign up with a confirmed email to get alerts, and every subscriber gets every company's drops at the same time. [1]

Model Odds is a web app that runs in the browser and needs JavaScript. [1] It is free, and so is its API, with no account needed. [1] The website and the API share one snapshot that is cached for up to five minutes, so results can be up to five minutes old. [1]

Other options include reading each company's own posts yourself, or running a general news search for a model name. Both put the checking in your hands, so decide how much of that work you want to do.

Common mistakes to avoid

Treating an estimate as a promised release date is a common mistake. An estimated chance tells you how likely a launch is, not when it will happen.

Trying a new model the day it appears without a prepared test makes it harder to compare results fairly. Keep the test the same from one model to the next.

Ignoring the way an estimate is built makes it harder to judge when to trust it. Look at how it was made before you act on it.

Relying on a single source for timing can leave gaps. Compare more than one view before you decide.

Forgetting to recheck your list of models means your watch may cover the wrong families. Update the list when your interests change.

Frequently asked questions

Is Model Odds free to use?

Yes. Model Odds is free, and so is its API, with no account needed. [1] It is a web app that runs in the browser and needs JavaScript turned on, so you open it in a browser tab. [1] It is also a website with a public API that serves per-line odds, the launch record and service health. [1]

Does Model Odds send alerts when a new model may be close?

You can sign up with a confirmed email to get alerts. [1] Every subscriber gets every company's drops at the same time, so the alerts do not depend on which company you follow. [1] The output is an estimated chance of release, not a fixed date, and it ranks model families by that chance within 30 days. [1]

Can Model Odds show how its estimates were made?

It runs three estimators side by side and shows the spread between them as an error bar. [1] The estimate is built from the gaps between recent launches and takes into account how long the current wait has already lasted. [1] Each past launch used has to be confirmed by an official post and an independent timeline, and it is listed with its date. [1]

Can Model Odds replay past dates to check its estimates?

The estimator can replay any past date without using data from after that date. [1] Signals read automatically from company developer kits and docs, and from third-party model catalogues when two of them agree, can only raise the odds. [1] Anthropic and OpenAI each have their own page that lists every past release, which helps you check the record yourself. [1]

Does Model Odds rank models from companies other than Anthropic and OpenAI?

Model Odds ranks every Anthropic and OpenAI model family by its estimated chance of a release within 30 days. [1] Anthropic and OpenAI each get their own company page. [1] For any other company, check the tool's own listing before you rely on it, and read how its estimates are built so you know what it covers.

Model Odds is free, and so is its API, with no account needed. [1]

Try Model Odds

Sources

  1. Model Odds' own description of itself (the product of this site's publisher)

Related comparisons