Skip to main content
Babbily gives you access to multiple models from one chat surface. The best model depends on what you are trying to do, how much context you need, and which capabilities are available for your plan. If you would rather not choose, select Auto at the top of the model picker. Auto picks a model and reasoning level for each message. See Auto model selection.
Model availability, names, and capabilities can change over time. Use the model picker as the source of truth for what is available in your account.

​
Choose by task

​
Current models

As of October 7, 2026, the model picker offers these 14 named models. The newest are Claude Haiku 5.5, Muse Spark 1.3, and DeepSeek V4.1 Flash, added October 7, and Claude Sonnet 5.5 and GPT-6.1 Sol, added October 6. Available tools and file limits depend on your selected model, plan, and account state. Every current model except Grok 4.7 has a context window of about 1 million tokens.

​
Capability signals

Hover over a model in the picker, or press and hold it on a touch screen, to see its description and the file types it accepts. Common capability differences include:
  • Context window size.
  • Tool availability.
  • Memory personalization availability.
  • Speed and reasoning depth.
  • Multi-step or agentic task behavior.
  • Codebase-scale context handling.
  • Fast mode availability.
  • API usage budget impact.
Every chat model accepts documents, text and code, images, audio, video, and ZIP archives, because Babbily reads what a model cannot read itself. See file attachments. Every chat model can also create images and videos with the built-in Image generation and Video generation tools. See image and video generation.

​
Sunsetting and removed models

No models are currently marked Sunsetting. On October 6 and 7, 2026, Babbily removed the models below from the picker, including every model that was previously marked Sunsetting. Saved chats keep their history, and older replies still show the model that wrote them. GPT Image 2.5 Flare and Grok Imagine Video left the picker as separate choices, but they still create your media: Image generation uses GPT Image 2.5 Flare, and Video generation uses Grok Imagine Video. See image and video generation. These are suggestions for new work, not promises of identical output. Recheck tools and API usage budget when you switch models.

​
What happens to chats that used a removed model

If the model selected in a chat was removed, Babbily selects another model and tells you once, for example:
  • Removed chat models switch to Auto.
  • If Auto is temporarily unavailable, Babbily selects your previous choice if it is still available, or the first model in the picker.
  • Nothing is resent automatically. If your last message did not get a reply, choose Retry to send it again with the new selection.
  • Babbily does not automatically pick the newer version of a removed model. Choose it from the picker if you want it.
Scheduled Tasks that used a removed model run on a replacement and say so. Open the task to choose a model yourself. See Scheduled Tasks.

​
Use Fast mode

Fast mode is available only when your selected model supports it. A lightning icon next to a model name means Fast mode is available for that model. Claude Opus 5.5, GPT-6.1 Sol, GPT-6 Astra, GPT-6 Luna, and GLM 5.3 Flash support Fast mode. Fast mode is a control on the selected model, not a separate picker choice, and the selected model name stays visible. Fast mode is not available while Auto is selected in the model picker. You can attach files in Fast mode. GLM 5.3 Flash in Fast mode does not read attachments itself, so Babbily reads them first and passes what it found. For work where visual detail matters, such as reading a chart or a screenshot closely, turn Fast mode off so GLM 5.3 Flash can see the image directly. “Flash” is part of some model names, not an indication that the model has this toggle. Gemini 3.8 Flash does not have a Fast mode control.
1

Select a supported model

Open the model picker and choose a model with the Fast mode lightning icon, such as Claude Opus 5.5 or GPT-6.1 Sol.
2

Turn on Fast mode

Choose Fast mode using the lightning control near the composer. Babbily shows a usage notice the first time you turn it on.
3

Send your prompt

Keep working with the selected model using its faster response path.
Fast mode can respond faster, but it uses more of your API usage budget. Babbily turns Fast mode off when you switch to a different model, including by opening a chat that uses a different model. Switching to Auto and back to the same model keeps your Fast mode setting, so check the control before your next prompt.

​
Choose reasoning effort

Open the composer menu, choose Manual, and turn on Reasoning. If the selected model supports effort levels, a Reasoning effort selector appears beside the microphone. These controls are hidden while Auto is selected in the model picker, because Auto sets the reasoning level for each reply. In Auto tool mode, Babbily chooses reasoning effort for supported work. When reasoning is off, either because tool mode is Off or because Reasoning is turned off in Manual, Babbily asks the model for its lightest reasoning setting. Some models stop reasoning entirely. Others, including Claude Opus 5.5, Claude Sonnet 5.5, GPT-6.1 Sol, GPT-6 Astra, Grok 4.7, Gemini 3.8 Flash, and Qwen 3.8 Max, have no fully-off setting and run at their lowest level instead. Higher effort can take longer and use more API usage budget. The available levels can change when you switch models.

​
Advanced reasoning and coding work

Some models are built for specialized advanced work, such as comparing multiple approaches, reviewing code, working across large files, or planning multi-step engineering tasks. Choose these models when the task benefits from depth, synthesis, or a larger working context. For better results:
  • State the goal, constraints, and expected output.
  • Include only the files or excerpts the model needs.
  • Ask for a plan before large edits or refactors.
  • Ask the model to call out uncertainty, risks, and tradeoffs.

​
Switch models during work

You can switch models when the task changes. Good switching patterns:
  • Use Auto when you would rather let Babbily match the model to each message.
  • Start with a faster model to brainstorm.
  • Switch to a stronger reasoning model for decisions or plans.
  • Turn on Image generation or Video generation for media work. You do not need a different model.
  • Use Fast mode when it appears and you want quicker supported replies, then check your API usage budget for heavier work.
  • Start a new thread if older context is steering the answer in the wrong direction.
When you switch models, restate the key goal if the thread is long. That helps the new reply focus on the current task.

​
Usage considerations

More advanced model work can use more of your API usage budget, especially when it includes tools, long file context, image generation, video generation, or deep research. For budget-sensitive work:
  • Use shorter prompts when the task is simple.
  • Avoid unnecessary file attachments.
  • Turn tools Off for model-only drafting.
  • Use search or deep research only when sources matter.
  • Check your usage card before running image or video generation.

​
Troubleshooting

If a capability is missing:
  • If Auto is selected, choose a named model to see Fast mode and reasoning controls.
  • Switch to a model that supports it.
  • Change tool mode from Off to Auto or Manual.
  • Check whether your plan supports the feature.
  • Check your remaining API usage budget.
  • Try a new thread if the current one has a long or confusing context history.

Auto model selection

Let Babbily choose a model and reasoning level for each message.

Known limits

Understand why capabilities can vary by model, plan, file, tool, and plugin.
Last modified on October 9, 2026