test

README.md

@@ -2,16 +2,30 @@
 frameworks:
 - Pytorch
 license: other
-tasks:
-- image-text-to-text
-language:
-- cn
-- en
+license_name: glm-4
+license_link: LICENSE
+pipeline_tag: image-text-to-text
+tags:
+  - glm
+  - edge
+inference: false
 ---
 
 # GLM-Edge-V-5B
 
-快速推理代码:
+中文阅读, 点击[这里](README_zh.md)
+
+## Inference with Transformers
+
+### Installation
+
+Install the transformers library from the source code:
+
+```shell
+pip install git+https://github.com/huggingface/transformers.git
+```
+
+### Inference
 
 ```python
 import torch
@@ -48,4 +62,8 @@ generate_kwargs = {
 output = model.generate(**generate_kwargs, max_new_tokens=100)
 print(tokenizer.decode(output[0][len(inputs["input_ids"][0]):], skip_special_tokens=True))
 
-```
+```
+
+## License
+
+The usage of this model’s weights is subject to the terms outlined in the [LICENSE](LICENSE).

README_zh.md

@@ -0,0 +1,53 @@
+# GLM-Edge-V-5B
+
+## 使用 transformers 库进行推理
+
+### 安装
+
+请安装源代码的transformers库。
+
+```shell
+pip install git+https://github.com/huggingface/transformers.git
+```
+### 推理
+
+```python
+import torch
+from PIL import Image
+from transformers import (
+    AutoTokenizer,
+    AutoImageProcessor,
+    AutoModelForCausalLM,
+)
+
+url = "img.png"
+messages = [{"role": "user", "content": [{"type": "image"}, {"type": "text", "text": "describe this image"}]}]
+image = Image.open(url)
+
+model_dir = "THUDM/glm-edge-v-5b"
+
+processor = AutoImageProcessor.from_pretrained(model_dir, trust_remote_code=True)
+tokenizer = AutoTokenizer.from_pretrained(model_dir, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(
+    model_dir,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    trust_remote_code=True,
+)
+
+inputs = tokenizer.apply_chat_template(
+    messages, add_generation_prompt=True, return_dict=True, tokenize=True, return_tensors="pt"
+).to(next(model.parameters()).device)
+
+generate_kwargs = {
+    **inputs,
+    "pixel_values": torch.tensor(processor(image).pixel_values).to(next(model.parameters()).device),
+}
+output = model.generate(**generate_kwargs, max_new_tokens=100)
+print(tokenizer.decode(output[0][len(inputs["input_ids"][0]):], skip_special_tokens=True))
+
+```
+
+## 协议
+
+本模型的权重的使用则需要遵循 [LICENSE](LICENSE)。

modeling_glm.py

@@ -31,7 +31,7 @@ from .configuration_glm import GlmConfig
 
 logger = logging.get_logger(__name__)
 
-_CHECKPOINT_FOR_DOC = "THUDM/glm-4-9b"
+_CHECKPOINT_FOR_DOC = "THUDM/glm-edge-5b"
 _CONFIG_FOR_DOC = "GlmConfig"