FreeWeights

Run LLMs locally. Chat never leaves your device.

Checking WebGPU

Resources

...

GPU access Checking...

0%

Storage used Checking...

≈

VRAM usage No model loaded

Model status Idle

Chat

Context usage estimate

Reply budget 0 Conversation 0

Older context compacted

Context nearly full

Conversation quality may degrade.

Selected model

Choose a model

Tags:

Previous chat found

Continue the saved local chat for this model, or start fresh and delete it.

Preparing model... 0%

Waiting for WebLLM progress reports.

Model Settings

Context window

Temperature Reply budget

Hardware

Select GPU for hardware-specific guidance

Memory amount

Select GPU first

Storage required -

VRAM required -

Params -

Default context -

Load failed

Technical details

Model Browser

Loading catalog...

⌄

Mobile-safe catalog Showing smaller models least likely to reset mobile browsers.

Search

Task

Size

Parameters

Quantization

Sort