The all-in-one local AI studio for your desktop: chat, image and video generation and a coding agent in one free, open source app. Windows and Linux. No Docker, no terminal, no cloud required.
TypeScript agentai-chatai-studioall-in-onecoding-agentdesktop-appfluximage-generationlinuxllmlocal-ailocal-llmopen-sourceprivacyself-hostedstable-diffusionuncensoreduncensored-aivideo-generationwindows
/**
* Which hosted models the LU Cloud strip in the chat picker shows, and in what
* order.
*
* Nebenbefund 3 of the R9 re-measure on the 2.6.7 Windows build (2026-08-30):
* in five looks at the picker, the LU CLOUD group showed five different groups
* of five, once "DeepSeek V4 Flash 0731, Qwen3 VL 30B, DeepSeek R1, Qwen 3.5
* 397B A17B, Kimi K3", shortly after "gpt-oss 120B, GLM 5, Hermes 3 405B, Qwen
* 3.6 27B, Qwen 3.6 35B A3B". It reads like a deliberately rotating taster.
*
* It is not one. Nothing in the app rotates, samples or shuffles that list:
* the strip took `.slice(0, 5)` off the models exactly as `/v1/models` handed
* them over, and that payload carries no rank, no "featured" flag and no
* stable order, so which five reach the screen is decided by whatever order
* the last answer happened to arrive in. The strip also never said the list
* was longer than five, so a model seen once looked simply gone.
*
* The answer is the smallest honest one: sort by the label the user reads, so
* the same five stand there every time, and name the number that did not fit.
*/
export const CLOUD_TEASER_LIMIT = 5
export interface TeaserModel {
name: string
displayName?: string
}
/** Deterministic, locale-independent: case-folded label, `name` breaks ties. */
function byLabel<T extends TeaserModel>(labelOf: (m: T) => string) {
return (a: T, b: T): number => {
const la = labelOf(a).toLowerCase()
const lb = labelOf(b).toLowerCase()
if (la !== lb) return la < lb ? -1 : 1
if (a.name !== b.name) return a.name < b.name ? -1 : 1
return 0
}
}
/**
* The rows the strip draws, and how many hosted models it left out.
* `models` is not mutated.
*/
export function cloudTeaserModels<T extends TeaserModel>(
models: T[],
labelOf: (m: T) => string = (m) => m.displayName ?? m.name,
limit: number = CLOUD_TEASER_LIMIT,
): { shown: T[]; more: number } {
const sorted = [...models].sort(byLabel(labelOf))
return {
shown: sorted.slice(0, limit),
more: Math.max(0, sorted.length - limit),
}
}