Update README.md
Browse files
README.md
CHANGED
|
@@ -101,10 +101,10 @@ vllm serve KennyVo/KAT-Coder-V2.5-Dev-OrnithVision --dtype float16
|
|
| 101 |
| ARC | ARC-Easy | 2376 | 99.1% | 98.9% | 99% |
|
| 102 |
| ARC | ARC-Challenge | 1172 | 96.9% | 98% | 97.1% |
|
| 103 |
| ARC | OVERALL | 3548 | 98.4% | 98.6% | 98.4% |
|
| 104 |
-
| MMLU | college_computer_science (STEM) | 100 | 97% | | |
|
| 105 |
-
| MMLU | computer_security (STEM) | 100 | 90% | | |
|
| 106 |
-
| MMLU | machine_learning (STEM) | 112 | 91.1% | | |
|
| 107 |
-
| MMLU | OVERALL | 312 | 92.6% | | |
|
| 108 |
|
| 109 |
### Vision benchmarks
|
| 110 |
|
|
|
|
| 101 |
| ARC | ARC-Easy | 2376 | 99.1% | 98.9% | 99% |
|
| 102 |
| ARC | ARC-Challenge | 1172 | 96.9% | 98% | 97.1% |
|
| 103 |
| ARC | OVERALL | 3548 | 98.4% | 98.6% | 98.4% |
|
| 104 |
+
| MMLU | college_computer_science (STEM) | 100 | 97% | | 93% |
|
| 105 |
+
| MMLU | computer_security (STEM) | 100 | 90% | | 88% |
|
| 106 |
+
| MMLU | machine_learning (STEM) | 112 | 91.1% | | 90.2% |
|
| 107 |
+
| MMLU | OVERALL | 312 | 92.6% | | 90.4% |
|
| 108 |
|
| 109 |
### Vision benchmarks
|
| 110 |
|