Update README.md
Browse files
README.md
CHANGED
@@ -11,13 +11,29 @@ base_model:
|
|
11 |
pipeline_tag: visual-question-answering
|
12 |
---
|
13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
<h1>Multi-modal Agent Tuning: Building a VLM-Driven Agent for Efficient Tool Usage</h1>
|
15 |
|
16 |
[GitHub](https://github.com/mat-agent/MAT-Agent.git) | [Project](https://mat-agent.github.io/)</a>
|
17 |
|
18 |
## MAT-MiniCPM-V 2.6
|
19 |
|
20 |
-
This model is a fine-tuned version of the [MiniCPM V2.6 7B](https://huggingface.co/openbmb/MiniCPM-V-2_6) model on the MM-traj dataset. On GTA and GAIA benchmarks, it achieved improvements of
|
21 |
|
22 |
## Usage
|
23 |
Our model inherits the inference architecture from [MiniCPM-V-2.6](https://huggingface.co/openbmb/MiniCPM-V-2_6). The following implementation is adapted from their original inference code with full compatibility.
|
@@ -157,13 +173,13 @@ print(answer)
|
|
157 |
## License
|
158 |
|
159 |
#### Model License
|
160 |
-
- The code in this repository is licensed under the
|
161 |
-
- Usage of our fine-tuned MiniCPM-based model weights must strictly adhere to the
|
162 |
|
163 |
#### Usage Terms
|
164 |
- β**Academic Research**: The model weights are freely available for academic use without restrictions.
|
165 |
- β**Commercial Use**:
|
166 |
-
- After completing the official
|
167 |
- Commercial users must maintain compliance with all terms outlined in the MiniCPM Model License.
|
168 |
|
169 |
#### Inheritance Clause
|
|
|
11 |
pipeline_tag: visual-question-answering
|
12 |
---
|
13 |
|
14 |
+
---
|
15 |
+
pipeline_tag: image-text-to-text
|
16 |
+
datasets:
|
17 |
+
- openbmb/RLAIF-V-Dataset
|
18 |
+
library_name: transformers
|
19 |
+
language:
|
20 |
+
- multilingual
|
21 |
+
tags:
|
22 |
+
- minicpm-v
|
23 |
+
- vision
|
24 |
+
- ocr
|
25 |
+
- multi-image
|
26 |
+
- video
|
27 |
+
- custom_code
|
28 |
+
---
|
29 |
+
|
30 |
<h1>Multi-modal Agent Tuning: Building a VLM-Driven Agent for Efficient Tool Usage</h1>
|
31 |
|
32 |
[GitHub](https://github.com/mat-agent/MAT-Agent.git) | [Project](https://mat-agent.github.io/)</a>
|
33 |
|
34 |
## MAT-MiniCPM-V 2.6
|
35 |
|
36 |
+
This model is a fine-tuned version of the [MiniCPM V2.6 7B](https://huggingface.co/openbmb/MiniCPM-V-2_6) model on the MM-traj dataset. On GTA and GAIA benchmarks, it achieved improvements of β18.59% and β7.78%β respectively compared to the non-fine-tuned baseline.
|
37 |
|
38 |
## Usage
|
39 |
Our model inherits the inference architecture from [MiniCPM-V-2.6](https://huggingface.co/openbmb/MiniCPM-V-2_6). The following implementation is adapted from their original inference code with full compatibility.
|
|
|
173 |
## License
|
174 |
|
175 |
#### Model License
|
176 |
+
- The code in this repository is licensed under the β[Apache-2.0 License](https://github.com/OpenBMB/MiniCPM/blob/main/LICENSE).
|
177 |
+
- Usage of our fine-tuned MiniCPM-based model weights must strictly adhere to the β[MiniCPM Model License](https://github.com/OpenBMB/MiniCPM/blob/main/MiniCPM%20Model%20License.md).
|
178 |
|
179 |
#### Usage Terms
|
180 |
- β**Academic Research**: The model weights are freely available for academic use without restrictions.
|
181 |
- β**Commercial Use**:
|
182 |
+
- After completing the official β[registration questionnaire](https://modelbest.feishu.cn/share/base/form/shrcnpV5ZT9EJ6xYjh3Kx0J6v8g)β and obtaining authorization, the MiniCPM-V 2.6 based weights (including our fine-tuned version) are available for commercial use free of charge.
|
183 |
- Commercial users must maintain compliance with all terms outlined in the MiniCPM Model License.
|
184 |
|
185 |
#### Inheritance Clause
|