unsloth/studio/backend/core/inference
Daniel Han 8ccb461570 studio: group GGUF shards by variant in size-based fallback
The smallest-fitting-variant fallback now groups split GGUF shards
by their variant prefix and sums all shard sizes per variant.

For example, DeepSeek-V3.2 UD-Q4_K_XL has 9 shards totaling
379.8 GB. The previous code treated each shard as a separate
"variant" and would have incorrectly selected a single 50 GB shard
as fitting, ignoring the other 8 shards needed.

Tested with unsloth/DeepSeek-V3.2-GGUF (237 GGUF files, 27
variants from 150 GB to 1.25 TB). Correctly groups and sorts
all variants by total size.
2026-03-15 05:24:06 -07:00
..
__init__.py Final cleanup 2026-03-12 18:28:04 +00:00
audio_codecs.py Final cleanup 2026-03-12 18:28:04 +00:00
inference.py Fix: Compare Mode Deadlock, Cancel Event Poisoning & IPC Optimization (#4303) 2026-03-15 16:11:44 +04:00
llama_cpp.py studio: group GGUF shards by variant in size-based fallback 2026-03-15 05:24:06 -07:00
orchestrator.py Fix: Compare Mode Deadlock, Cancel Event Poisoning & IPC Optimization (#4303) 2026-03-15 16:11:44 +04:00
worker.py Final cleanup 2026-03-12 18:28:04 +00:00