Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.
Python agentaichatgptdeepseekfine-tuninggemmaimage-generationllamallmllmsopenaipythonqwenreinforcement-learningself-hostedstable-diffusiontext-to-speechttsuiunsloth
// SPDX-License-Identifier: AGPL-3.0-only
// Copyright 2026-present the Unsloth AI Inc. team. All rights reserved. See /studio/LICENSE.AGPL-3.0
import type { DiffusionSampleImage } from "../api";
/** The backend refuses more than this many sample prompts. */
export const MAX_SAMPLE_PROMPTS = 4;
const LINE_BREAK = /\r?\n/;
export function parseSamplePrompts(text: string): string[] {
return text
.split(LINE_BREAK)
.map((line) => line.trim())
.filter((line) => line.length > 0)
.slice(0, MAX_SAMPLE_PROMPTS);
}
export interface SampleRound {
step: number;
images: DiffusionSampleImage[];
}
/** Preview images grouped into rounds by step, oldest first; within a round in prompt order. */
export function groupSamplesByStep(
samples: DiffusionSampleImage[] | null | undefined,
): SampleRound[] {
const byStep = new Map<number, DiffusionSampleImage[]>();
for (const s of samples ?? []) {
if (!s || typeof s.path !== "string" || !Number.isFinite(s.step)) {
continue;
}
const list = byStep.get(s.step) ?? [];
list.push(s);
byStep.set(s.step, list);
}
return Array.from(byStep.entries())
.sort((a, b) => a[0] - b[0])
.map(([step, images]) => ({ step, images }));
}