Update README.md
Browse files
README.md
CHANGED
@@ -44,11 +44,10 @@ FlagEval (Libra)** is a comprehensive evaluation system and open platform for la
|
|
44 |
|
45 |
| Metrics | Qwen3-235B-A22B-Instruct-2507-H100-CUDA | Qwen3-235B-A22B-Instruct-2507-FlagOS |
|
46 |
| --------- | ------------------ | ---------------------- |
|
47 |
-
|
|
48 |
-
| AIME |
|
49 |
-
|
|
50 |
-
|
|
51 |
-
| MUSR | Coming soon | Coming soon |
|
52 |
|
53 |
# User Guide
|
54 |
|
|
|
44 |
|
45 |
| Metrics | Qwen3-235B-A22B-Instruct-2507-H100-CUDA | Qwen3-235B-A22B-Instruct-2507-FlagOS |
|
46 |
| --------- | ------------------ | ---------------------- |
|
47 |
+
| liveBench | 0.753 | 0.752 |
|
48 |
+
| AIME | 0.833 | 0.800 |
|
49 |
+
| MMLU | 0.833 | 0.835 |
|
50 |
+
| MUSR | 0.597 | 0.615 |
|
|
|
51 |
|
52 |
# User Guide
|
53 |
|