ahmedheakl
commited on
Commit
•
095b85c
1
Parent(s):
3a7d20d
Update README.md
Browse files
README.md
CHANGED
@@ -140,12 +140,11 @@ print(converted_risc)
|
|
140 |
|----------------------------------------|-------------------------------|---------------------|-----------------------|
|
141 |
| GPT4o | 1293 | 0% | 7.55% |
|
142 |
| DeepSeekCoder2-16B | 1483 | 0% | 6.29% |
|
143 |
-
|
144 |
-
| DeepSeekCoder-1.3B-xTokenizer-
|
145 |
-
| DeepSeekCoder-1.3B-xTokenizer
|
146 |
-
| DeepSeekCoder-1.3B-xTokenizer | **27** | **69.81%** | **88.68%** |
|
147 |
|
148 |
-
**Table:** Comparison of models' performance on the _x86 to RISCv64_ transpilation task.
|
149 |
|
150 |
|
151 |
|
|
|
140 |
|----------------------------------------|-------------------------------|---------------------|-----------------------|
|
141 |
| GPT4o | 1293 | 0% | 7.55% |
|
142 |
| DeepSeekCoder2-16B | 1483 | 0% | 6.29% |
|
143 |
+
| **DeepSeekCoder-1.3B-xTokenizer-int4** | 112 | 14.47% | 68.55% |
|
144 |
+
| **DeepSeekCoder-1.3B-xTokenizer-int8** | 31 | 69.81% | 88.05% |
|
145 |
+
| **DeepSeekCoder-1.3B-xTokenizer** | **27** | **69.81%** | **88.68%** |
|
|
|
146 |
|
147 |
+
**Table:** Comparison of models' performance on the _x86 to RISCv64_ transpilation task. The top section lists pre-existing models, while the bottom section lists models trained by us.
|
148 |
|
149 |
|
150 |
|