Colorfle Solver

What Colorfle Is and Why It Is Harder Than It Looks

Colorfle is a daily puzzle where you see one target color and must find the three source colors that mix to produce it. The twist is that you pick from a fixed palette of about twenty colors, and the mixing uses weighted RGB averaging across both YCC and RGB color spaces. You have six guesses. Each guess shows three color swatches, and you get feedback on how close each one is to the answer. Green means exact match. Yellow means close but in the wrong position. Gray means the color is not in the answer at all.

Most people expect to solve Colorfle the same way they solve Wordle — just keep guessing until something clicks. That approach fails in Colorfle because the color space is large enough that brute-force guessing rarely works in six tries. There are over a thousand possible three-color permutations from a palette of twenty colors, and only six guesses to find the right one. Without a systematic elimination strategy, you are essentially guessing at odds of roughly 1 in 170, which is why most people run out of guesses before finding the answer.

Here is why Colorfle consistently trips people up: the game gives you feedback per color position, not per color. If your first color swatch is wrong, you mark it gray. If it is close, you mark it yellow. If it matches exactly, you mark it green. After two or three guesses with feedback, you have a dense web of constraints that is difficult to reason about mentally. The solver uses those constraints computationally to eliminate impossible combinations and show you what remains. This is exactly the kind of problem where computational help provides the most value.

The game also has a feature that catches new players off guard: different three-color combinations can produce the same or nearly the same mixed color. This is called "color degeneracy," and it means that even when you have narrowed your candidates to a handful of combinations, some of them may look identical on screen. You need feedback from the game itself, not just visual similarity, to distinguish between degenerate combinations. The solver accounts for this by ranking candidates by their actual RGB distance to the target, not by visual appearance alone.

Using the Solver: Hex-Solving vs. Feedback-Solving

Hex-solving works when you have a target color but no guesses yet. Enter the hex code into the input field, or use the color picker to select it visually from the game screen. Click Solve, and the solver finds the five three-color combinations that produce the closest match to your target. Each suggestion shows the three component colors, their mixing weights, and the resulting mixed hex code alongside the similarity score. You can click "Use This" on any suggestion to add it as a guess in your feedback panel.

Feedback-solving works after you have made guesses in the actual game and marked each color swatch with green, yellow, or gray. Click each swatch in the guess row to cycle through feedback states. Once all three swatches in a guess have feedback set, the Refine button becomes active. Click it and the solver removes every combination that would produce different feedback than what you marked. This elimination is mathematically precise: if a combination cannot produce the feedback you observed, it is removed from the candidate list. No approximations, no fuzzy matching.

The two modes work together seamlessly. Start with hex-solving to get initial candidates, then use feedback-solving to narrow them down. After two refinement cycles, you are usually looking at one or two combinations left. That is when you make your final guess with confidence. The recommended workflow is: enter the target hex, solve to get five candidates, pick the top candidate as your first guess in the game, observe the feedback, mark it in the solver, refine, pick the next candidate, repeat. Most puzzles are solvable in three guesses using this workflow.

If you prefer to make your own guesses rather than following the solver's suggestions, you can still use the feedback-solving mode. Enter any three colors as a guess, mark the feedback the game gives you, and refine. The solver does not care where your guesses come from — it only uses the feedback pattern to eliminate candidates. This flexibility means you can combine your own intuition with the solver's elimination logic for a hybrid approach that plays to the strengths of both human judgment and computational precision.

What the Green, Yellow, and Gray Feedback Actually Means

Green means the color at that position matches exactly. The correct answer has the same color at the same position. This is the most useful feedback because it locks in one of the three slots permanently. Once you have a green, that position is solved and you never need to change it. The remaining work is figuring out the other two positions.

Yellow means the color exists somewhere in the answer but not at the position where you placed it. The color is part of the correct combination, but it belongs in a different slot. This is similar to Wordle's yellow feedback for letters. When you see yellow, you know two things: the color is definitely in the answer, and it is definitely not in the position you guessed. Use this information to rearrange your next guess accordingly.

Gray means the color does not appear in the answer at all. This is straightforward but powerful feedback. Every gray swatch eliminates that color from all three positions in every remaining candidate combination. After three guesses with gray feedback, you have typically eliminated 60-70% of the color palette, which dramatically shrinks the search space.

The feedback follows the same rules as Wordle's letter feedback system. If the answer is [Red, Blue, Yellow] and you guess [Blue, Red, Green], the feedback would be [yellow, yellow, gray]. Blue is in the answer but not at position one. Red is in the answer but not at position two. Green is not in the answer at all. Click each swatch to mark the correct feedback, then refine. Getting the feedback wrong is the most common reason the solver returns "no matches found." If you mark a swatch green when it should be yellow, the solver eliminates the correct answer because no valid combination can be both green and yellow at the same position. Always double-check your feedback markings before clicking Refine.

Why Your Target Might Show Multiple Plausible Combinations

The color space in Colorfle is quantized. You pick from twenty specific colors, and mixing is done with weighted averaging. This means different three-color combinations can and do produce the same, or nearly the same, target color. The solver calls this "color degeneracy," and it is not a flaw in the game or the solver — it is an inherent property of discrete color mixing. When two combinations produce visually identical results, no amount of visual inspection can tell you which one is the official answer. Only feedback from the game itself can resolve the ambiguity.

When you enter a hex and see five suggestions that all look reasonable, that is the degeneracy at work. The mixed result of combination A might be #7a9c3e and combination B might be #7b9d3f. These are visually identical on any screen. The similarity score tells you how close each combination gets to the target, but it cannot tell you which one is the official answer. You narrow that down through feedback, not through similarity alone. The solver ranks candidates by similarity to help you prioritize, but the final determination always comes from the game's feedback.

Mid-range targets — colors that are neither very dark nor very saturated — tend to have the most degeneracy. This is because mixing is essentially weighted averaging, and averages of different sets of numbers can converge to the same result. Deep greens and bright reds tend to be more distinctive because they occupy extreme positions in the color space where fewer combinations can reach. If you are stuck between two suggestions, make one your next guess and pay attention to the feedback pattern. Small differences in feedback — even a single yellow vs. gray on one position — will quickly separate the remaining candidates.