Spaces:
Sleeping
Sleeping
Commit History
-
Fix SVG download: use single quotes inside style attr, no duplicate style
Verified signature964b628 -
Cap to top 30 categories, hide gr.File X button, single-row buttons, no duplicate title
Verified signature7d5a424 -
Result narration (Qwen-powered) + UI polish for upload area + dark-mode fixes
Verified signature5ae5227 -
Result narration (Qwen-powered) + UI polish for upload area + dark-mode fixes
Verified signature7af78a0 -
SQL self-correction: 3 retries with error context + GROUP BY rule in prompt
Verified signaturef5b7f67 -
SQL self-correction: 3 retries with error context + GROUP BY rule in prompt
Verified signaturee6681c9 -
SVG: standalone download (XML prolog, white bg, fixed dims) + Plotly first
Verified signature0ac4ef2 -
SVG: standalone download (XML prolog, white bg, fixed dims) + Plotly first
Verified signature89360e0 -
Better SQL prompt: column-name mapping rules + few-shot + distinct value hints
Verified signaturea7fa6d3 -
Better SQL prompt: column-name mapping rules + few-shot + distinct value hints
Verified signatureb0a300f
-
Pin kaleido==0.2.1 + native SVG fallback when kaleido fails
Verified signature064c0d2 -
UX: CSV/SVG download links, schema preview on load, cleaner SVG theme
Verified signaturea172c0c -
Load LoRA via PeftModel on top of standard base models (fixes r=16 vs r=8 mismatch)
Verified signature1bbdff9 -
Load LoRA via PeftModel on top of standard base models (fixes r=16 vs r=8 mismatch)
Verified signatureeb30a86 -
Load LoRA via PeftModel on top of standard base models (fixes r=16 vs r=8 mismatch)
Verified signature730b25d -
ZeroGPU best practice: load models at module level (cuda), inference only inside @spaces.GPU
Verified signature420b1db -
ZeroGPU best practice: load models at module level (cuda), inference only inside @spaces.GPU
Verified signature61aee8d -
ZeroGPU best practice: load models at module level (cuda), inference only inside @spaces.GPU
Verified signaturea57eca6 -
ZeroGPU best practice: load models at module level (cuda), inference only inside @spaces.GPU
Verified signaturec2ac226 -
Sequential model load/unload: keep peak VRAM at ~14GB
Verified signature292fdf4 -
Sequential model load/unload: keep peak VRAM at ~14GB
Verified signaturefab3ba1 -
Initial deploy: Apple/Claude design, DuckDB, 3 trained LoRAs
Verified signaturea067ada