Drop 'demo' suffix from page title
Browse files
README.md
CHANGED
|
@@ -9,7 +9,7 @@ pinned: false
|
|
| 9 |
license: apache-2.0
|
| 10 |
---
|
| 11 |
|
| 12 |
-
# KakeyaLattice KV-cache compression
|
| 13 |
|
| 14 |
Side-by-side comparison of **bf16 DynamicCache** vs **KakeyaLattice E8**
|
| 15 |
compression at three quality levels (Q=10 aggressive, Q=38 balanced,
|
|
|
|
| 9 |
license: apache-2.0
|
| 10 |
---
|
| 11 |
|
| 12 |
+
# KakeyaLattice KV-cache compression
|
| 13 |
|
| 14 |
Side-by-side comparison of **bf16 DynamicCache** vs **KakeyaLattice E8**
|
| 15 |
compression at three quality levels (Q=10 aggressive, Q=38 balanced,
|
app.py
CHANGED
|
@@ -146,9 +146,9 @@ EXAMPLE_PROMPTS = [
|
|
| 146 |
]
|
| 147 |
|
| 148 |
|
| 149 |
-
with gr.Blocks(title="KakeyaLattice KV-cache compression
|
| 150 |
gr.Markdown(
|
| 151 |
-
"# KakeyaLattice KV-cache compression
|
| 152 |
"Compare generation output + latency across **bf16 baseline** and "
|
| 153 |
"three **KakeyaLattice E8** compression levels on a small HF causal LM. "
|
| 154 |
"The E8 variant uses 8-D nested-lattice closest-point quantisation "
|
|
|
|
| 146 |
]
|
| 147 |
|
| 148 |
|
| 149 |
+
with gr.Blocks(title="KakeyaLattice KV-cache compression") as demo:
|
| 150 |
gr.Markdown(
|
| 151 |
+
"# KakeyaLattice KV-cache compression\n\n"
|
| 152 |
"Compare generation output + latency across **bf16 baseline** and "
|
| 153 |
"three **KakeyaLattice E8** compression levels on a small HF causal LM. "
|
| 154 |
"The E8 variant uses 8-D nested-lattice closest-point quantisation "
|