zhiyucheng commited on
Commit
e560322
·
1 Parent(s): f38b5ad

update model card

Browse files
Files changed (1) hide show
  1. README.md +2 -3
README.md CHANGED
@@ -30,7 +30,6 @@ GOVERNING TERMS: Use of this model is governed by the [NVIDIA Open Models Licens
30
  ## Software Integration
31
  **Supported Runtime Engine(s):** <br>
32
  * Tensor(RT)-LLM <br>
33
- * vLLM <br>
34
 
35
  **Supported Hardware Microarchitecture Compatibility:** <br>
36
  * NVIDIA Blackwell <br>
@@ -41,7 +40,7 @@ GOVERNING TERMS: Use of this model is governed by the [NVIDIA Open Models Licens
41
  * Linux <br>
42
 
43
  ## Model Version(s):
44
- v0.17.0 <br>
45
 
46
  # Training and Evaluation Datasets:
47
 
@@ -157,7 +156,7 @@ The accuracy (MMLU, 5-shot) and Medusa acceptance rate benchmark results are pre
157
  | FP8 | 68.3 | 2.07 |
158
 
159
  ## Inference:
160
- **Engine:** Tensor(RT)-LLM or vLLM <br>
161
  **Test Hardware:** H100 <br>
162
 
163
  ## Ethical Considerations
 
30
  ## Software Integration
31
  **Supported Runtime Engine(s):** <br>
32
  * Tensor(RT)-LLM <br>
 
33
 
34
  **Supported Hardware Microarchitecture Compatibility:** <br>
35
  * NVIDIA Blackwell <br>
 
40
  * Linux <br>
41
 
42
  ## Model Version(s):
43
+ v0.21.0 <br>
44
 
45
  # Training and Evaluation Datasets:
46
 
 
156
  | FP8 | 68.3 | 2.07 |
157
 
158
  ## Inference:
159
+ **Engine:** Tensor(RT)-LLM <br>
160
  **Test Hardware:** H100 <br>
161
 
162
  ## Ethical Considerations