Brighten (HSV)

The README nails the difference in one sentence: brighten green in HSV and you get bright green; lighten green in HSL and you get light green - which is to say, closer to white. Brighten (HSV) is the version of the pack's lighten tool that boosts the value channel without washing color out toward gray. If you've ever lifted an image and thought "why does everything look faded now," this is the node you wanted instead. It's the direct companion to the pack's Lighten (HSL), and once you've felt the difference you'll reach for it constantly.

How it works

The node converts the image to HSV, adds the factor to the value channel, clamps to the valid range, and converts back to RGB. Since it edits only value, hue and saturation are untouched - a saturated red stays a saturated red as it brightens, it just becomes a brighter saturated red. Contrast that with HSL Lighten, which adds to lightness and drags colors toward white as it lifts them. Same input, same output shape, very different look, and the author's example image showing both side by side is worth a look before you pick.

The inputs that matter

• image - any IMAGE tensor.
• factor (default 0.5, range −1 to 1) - the amount added to value. 0 is no change; 0.5 lifts pure black toward 50% brightness; 1 pushes everything to full value while preserving hue and saturation; negative values darken instead.
• masked_area + optional mask_opt - the pack's standard masking.

Output: one image tensor, same size as the input.

A couple of honest notes

The in-UI description is its usual unreliable self - it claims factor 0 gives a "completely bright and saturated image" and factor 1 gives the original, which is backwards. Code says value += factor; 0 is original, higher is brighter. Trust the slider, not the tooltip.

The other thing to know: at high factors, brightening in HSV keeps colors saturated, which is great - but it can also start to look neon, because nothing is washing out. A factor of 1.0 on an already-vivid image is aggressive. For most corrections you'll live in the 0.2–0.6 band.

Installing it

ComfyUI Manager → search "Basix Image Filters" (repo basix_image_filters) → install → restart. Or by hand:

cd ComfyUI/custom_nodes
git clone https://github.com/maludwig/basix_image_filters

Restart, done. No models, no dependencies beyond stock torch.

Putting it to work

The standard move is the img2img hint: brighten a reference image, run it back through the sampler at 0.7–0.9 denoise, and the model reproduces the brighter, more colorful tone instead of re-muddying it. If only part of the frame should glow - the window, not the wall - use mask_opt with a MaskEditor mask (right-click a Preview Image node → Copy (Clipspace) → paste into a Load Image node → Open in MaskEditor) and flip masked_area to pick which side of the mask gets the boost. Same mask can be reused for the inverse effect by toggling it off; the pack's whole masking trick in one node.