Upload mla_absorbed_cache_report_en.md with huggingface_hub
Browse files
mla_absorbed_cache_report_en.md
CHANGED
|
@@ -194,4 +194,4 @@ This report validates and implements **complete VRAM optimization for DeepSeek-V
|
|
| 194 |
- `l40s_quant.py` — per-channel/per-tensor quantization comparison
|
| 195 |
- `l40s_opt3.py` — deep optimization (k_pe quantization/mixed precision/INT4)
|
| 196 |
|
| 197 |
-
*Cloud LTE Studio · 2026-08-08 ·
|
|
|
|
| 194 |
- `l40s_quant.py` — per-channel/per-tensor quantization comparison
|
| 195 |
- `l40s_opt3.py` — deep optimization (k_pe quantization/mixed precision/INT4)
|
| 196 |
|
| 197 |
+
*Cloud LTE Studio · 2026-08-08 · GPL-3.0 License*
|