Commit
·
1a107f0
1
Parent(s):
ce9674f
Upload visit_bench_leaderboard.tsv
Browse files- visit_bench_leaderboard.tsv +12 -0
visit_bench_leaderboard.tsv
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
Model RFE Battles Win vs. Verified GPT-4
|
| 2 |
+
Human Verified GPT-4 Reference 1363 3274 -
|
| 3 |
+
LLaVA (13B) 1099 3274 5.03%
|
| 4 |
+
mPLUG-Owl (7B) 1053 3284 4.55%
|
| 5 |
+
LlamaAdapter-v2 (7B) 1037 3281 3.8%
|
| 6 |
+
Otter (9B) 998 154 2.50%
|
| 7 |
+
InstructBLIP (13B) 992 3274 2.37%
|
| 8 |
+
VisualGPT (Da Vinci 003) 967 251 1.92%
|
| 9 |
+
MiniGPT-4 (7B) 925 3291 2.09%
|
| 10 |
+
OpenFlamingo (9B) 892 441 0.0%
|
| 11 |
+
Multimodal GPT 854 267 0.0%
|
| 12 |
+
PandaGPT (13B) 820 3275 0.85%
|