wip

matatonic · Jul 15, 2024 · 72b248e · 72b248e
1 parent 0e5d945
commit 72b248e
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 20 deletions.
diff --git a/README.md b/README.md
@@ -10,7 +10,7 @@ An OpenAI API compatible vision server, it functions like `gpt-4-vision-preview`
 ## Model support
 
 - [X] [OpenGVLab](https://huggingface.co/OpenGVLab)
-- - [ ] [InternVL2-40B](https://huggingface.co/OpenGVLab/InternVL2-40B) (currently errors)
+- - [X] [InternVL2-40B](https://huggingface.co/OpenGVLab/InternVL2-40B)
 - - [X] [InternVL2-26B](https://huggingface.co/OpenGVLab/InternVL2-26B)
 - - [X] [InternVL2-8B](https://huggingface.co/OpenGVLab/InternVL2-8B)
 - - [X] [InternVL2-4B](https://huggingface.co/OpenGVLab/InternVL2-4B) (alternate docker only)
@@ -49,7 +49,7 @@ An OpenAI API compatible vision server, it functions like `gpt-4-vision-preview`
 - - [X] [360VL-8B](https://huggingface.co/qihoo360/360VL-8B)
 - - [X] [360VL-70B](https://huggingface.co/qihoo360/360VL-70B) (untested)
 - [X] [LlavaNext](https://huggingface.co/llava-hf)
-- - [X] [llava-v1.6-34b-hf](https://huggingface.co/llava-hf/llava-v1.6-34b-hf) (currently errors, use an image before 0.26.0)
+- - [X] [llava-v1.6-34b-hf](https://huggingface.co/llava-hf/llava-v1.6-34b-hf)
 - - [X] [llava-v1.6-vicuna-13b-hf](https://huggingface.co/llava-hf/llava-v1.6-vicuna-13b-hf)
 - - [X] [llava-v1.6-vicuna-7b-hf](https://huggingface.co/llava-hf/llava-v1.6-vicuna-7b-hf)
 - - [X] [llava-v1.6-mistral-7b-hf](https://huggingface.co/llava-hf/llava-v1.6-mistral-7b-hf)
@@ -115,8 +115,7 @@ See: [OpenVLM Leaderboard](https://huggingface.co/spaces/opencompass/open_vlm_le
 
 Version 0.27.0
 
-- new model support: OpenGVLab/InternVL2 series of models (40B still has errors, 4B requires alternate docker image)
-- 
+- new model support: OpenGVLab/InternVL2 series of models (1B, 2B, 4B, 8B*, 26B*, 40B*) - *(current top open source models)
 
 Version 0.26.0
 

diff --git a/model_conf_tests.alt.json b/model_conf_tests.alt.json
@@ -15,9 +15,5 @@
   ["YanweiLi/MGM-8x7B", "--use-flash-attn"],
   ["YanweiLi/MGM-8x7B-HD", "--load-in-4bit", "--use-flash-attn"],
   ["YanweiLi/MGM-8x7B-HD", "--use-flash-attn"],
-  ["internlm/internlm-xcomposer2-4khd-7b", "--use-flash-attn", "--device-map", "cuda:0", "--load-in-4bit"],
-  ["internlm/internlm-xcomposer2-7b", "--use-flash-attn", "--device-map", "cuda:0", "--load-in-4bit"],
-  ["internlm/internlm-xcomposer2-vl-1_8b", "--use-flash-attn", "--device-map", "cuda:0", "--load-in-4bit"],
-  ["internlm/internlm-xcomposer2-vl-7b", "--use-flash-attn", "--device-map", "cuda:0", "--load-in-4bit"],
   ["vikhyatk/moondream1"]
 ]
diff --git a/model_conf_tests.json b/model_conf_tests.json
@@ -25,18 +25,18 @@
   ["OpenGVLab/InternVL-Chat-V1-5", "--device-map", "cuda:0", "--max-tiles", "40"],
   ["OpenGVLab/InternVL-Chat-V1-5", "--device-map", "cuda:0"],
   ["OpenGVLab/InternVL-Chat-V1-5-Int8", "--device-map", "cuda:0"],
-  ["OpenGVLab/InternVL2-1B", "--load-in-4bit"],
-  ["OpenGVLab/InternVL2-1B"],
-  ["OpenGVLab/InternVL2-2B", "--load-in-4bit"],
-  ["OpenGVLab/InternVL2-2B"],
-  ["OpenGVLab/InternVL2-4B", "--load-in-4bit"],
-  ["OpenGVLab/InternVL2-4B"],
-  ["OpenGVLab/InternVL2-8B", "--load-in-4bit"],
-  ["OpenGVLab/InternVL2-8B"],
-  ["OpenGVLab/InternVL2-26B", "--load-in-4bit"],
-  ["OpenGVLab/InternVL2-26B"],
-  ["OpenGVLab/InternVL2-40B", "--load-in-4bit"],
-  ["OpenGVLab/InternVL2-40B"],
+  ["OpenGVLab/InternVL2-1B", "--device-map", "cuda:0", "--load-in-4bit"],
+  ["OpenGVLab/InternVL2-1B", "--device-map", "cuda:0"],
+  ["OpenGVLab/InternVL2-2B", "--device-map", "cuda:0", "--load-in-4bit"],
+  ["OpenGVLab/InternVL2-2B", "--device-map", "cuda:0"],
+  ["OpenGVLab/InternVL2-4B", "--device-map", "cuda:0", "--load-in-4bit"],
+  ["OpenGVLab/InternVL2-4B", "--device-map", "cuda:0"],
+  ["OpenGVLab/InternVL2-8B", "--device-map", "cuda:0", "--load-in-4bit"],
+  ["OpenGVLab/InternVL2-8B", "--device-map", "cuda:0"],
+  ["OpenGVLab/InternVL2-26B", "--device-map", "cuda:0", "--load-in-4bit"],
+  ["OpenGVLab/InternVL2-26B", "--device-map", "cuda:0"],
+  ["OpenGVLab/InternVL2-40B", "--device-map", "cuda:0", "--load-in-4bit"],
+  ["OpenGVLab/InternVL2-40B", "--device-map", "cuda:0"],
   ["OpenGVLab/Mini-InternVL-Chat-2B-V1-5", "--load-in-4bit"],
   ["OpenGVLab/Mini-InternVL-Chat-2B-V1-5", "--max-tiles", "40", "--load-in-4bit"],
   ["OpenGVLab/Mini-InternVL-Chat-2B-V1-5", "--max-tiles", "40"],