faxenoff commited on
Commit
79db0e8
·
verified ·
1 Parent(s): d0a4180

card: neutral what-is-in-repo wording

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -91,9 +91,9 @@ def embed(texts, max_len=128, mrl_dim=768):
91
 
92
  ## What's in this repo — ready-to-run compiled engines
93
 
94
- This repo is the model's **CDN**: it holds **pre-compiled, ready-to-run engines**, named per
95
- **runtime × GPU arch × OS × length-bucket** so the app downloads only the file matching the user's
96
- hardware. No compilation on the user's machine.
97
 
98
  - **TensorRT** `*.engine` — NVIDIA, INT8 W8A16, per arch × OS × bucket:
99
  `code-daemon-embed-v1-{s,m,l}_{win_x64,linux_x64}_trt_sm_{86,89,120}.engine`
 
91
 
92
  ## What's in this repo — ready-to-run compiled engines
93
 
94
+ This repo holds **pre-compiled, ready-to-run engines**, named per
95
+ **runtime × GPU arch × OS × length-bucket** grab the compiled model that matches your runtime and
96
+ hardware and use it directly, with no compilation on your machine.
97
 
98
  - **TensorRT** `*.engine` — NVIDIA, INT8 W8A16, per arch × OS × bucket:
99
  `code-daemon-embed-v1-{s,m,l}_{win_x64,linux_x64}_trt_sm_{86,89,120}.engine`