ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Thread the task-scoped GGUF fit budget through every picker expander

Browse source 
The previous commit introduced the single-device budget but wired it
into only the downloaded-group and Hub search sites. The LM Studio,
custom-folder, local-dir, live-search and exported-GGUF expanders
reachable from the Images/Video pickers still measured against the
summed multi-GPU total, as did the size-based GGUF row badge, so those
paths could still recommend a quant that OOMs on a single device. All
GgufVariantExpander call sites in HubModelPicker now share
expanderGpuGb, and the row badge derives the same task-scoped budget.
This commit is contained in:
Daniel Han 2026-07-11 18:36:31 +00:00
commit dcd6666f8d
1 changed files with 11 additions and 25 deletions
Download patch file Download diff file Expand all files Collapse all files

36
studio/frontend/src/components/assistant-ui/model-selector/pickers.tsx
View file

@ -2217,7 +2217,9 @@ export function HubModelPicker({
sizeBytes != null &&
!fitsDevice({
sizeBytes,
gpuGb: gpu.memoryTotalGb,
// Same task-scoped budget as expanderGpuGb: single device for
// Images/Video (one-device pipelines), summed total for chat.
gpuGb: task ? gpu.maxDeviceMemoryGb : gpu.memoryTotalGb,
systemRamGb: gpu.systemRamAvailableGb,
});
map.set(r.id, {
@ -3989,9 +3991,7 @@ export function HubModelPicker({
onNavigatePastEnd={() =>
hubModelList.moveFocus(optionKey, "next")
}
gpuGb={
gpu.available ? gpu.memoryTotalGb : undefined
}
gpuGb={expanderGpuGb}
systemRamGb={gpu.systemRamAvailableGb || undefined}
/>
)}
@ -4080,9 +4080,7 @@ export function HubModelPicker({
onNavigatePastEnd={() =>
hubModelList.moveFocus(optionKey, "next")
}
gpuGb={
gpu.available ? gpu.memoryTotalGb : undefined
}
gpuGb={expanderGpuGb}
systemRamGb={gpu.systemRamAvailableGb || undefined}
/>
)}
@ -4161,9 +4159,7 @@ export function HubModelPicker({
onNavigatePastEnd={() =>
hubModelList.moveFocus(optionKey, "next")
}
gpuGb={
gpu.available ? gpu.memoryTotalGb : undefined
}
gpuGb={expanderGpuGb}
systemRamGb={gpu.systemRamAvailableGb || undefined}
/>
)}
@ -4261,9 +4257,7 @@ export function HubModelPicker({
}}
vramStatus={info?.status ?? null}
vramEst={info?.est}
gpuGb={
gpu.available ? gpu.memoryTotalGb : undefined
}
gpuGb={expanderGpuGb}
onArrowDownIntoChildren={
expandedGguf === id
? () => focusFirstChildOption(optionKey)
@ -4282,9 +4276,7 @@ export function HubModelPicker({
onNavigatePastEnd={() =>
hubModelList.moveFocus(optionKey, "next")
}
gpuGb={
gpu.available ? gpu.memoryTotalGb : undefined
}
gpuGb={expanderGpuGb}
systemRamGb={gpu.systemRamAvailableGb || undefined}
variantActions={{
onDelete: async (quant) => {
@ -4375,9 +4367,7 @@ export function HubModelPicker({
onNavigatePastEnd={() =>
hubModelList.moveFocus(optionKey, "next")
}
gpuGb={
gpu.available ? gpu.memoryTotalGb : undefined
}
gpuGb={expanderGpuGb}
systemRamGb={gpu.systemRamAvailableGb || undefined}
variantActions={{
onDelete: async (quant) => {
@ -4441,9 +4431,7 @@ export function HubModelPicker({
isSearchGguf ? null : (vram?.status ?? null)
}
vramEst={isSearchGguf ? undefined : vram?.est}
gpuGb={
gpu.available ? gpu.memoryTotalGb : undefined
}
gpuGb={expanderGpuGb}
onArrowDownIntoChildren={
expandedGguf === id
? () => {
@ -4466,9 +4454,7 @@ export function HubModelPicker({
onNavigatePastEnd={() =>
hubModelList.moveFocus(optionKey, "next")
}
gpuGb={
gpu.available ? gpu.memoryTotalGb : undefined
}
gpuGb={expanderGpuGb}
systemRamGb={gpu.systemRamAvailableGb || undefined}
variantActions={{
onDelete: async (quant) => {