tc-mb
commited on
Commit
·
1fabcfe
1
Parent(s):
18ae5cb
update readme
Browse files
README.md
CHANGED
@@ -16,22 +16,22 @@ tags:
|
|
16 |
|
17 |
<h1>A GPT-4V Level MLLM for Single Image, Multi Image and Video on Your Phone</h1>
|
18 |
|
19 |
-
[GitHub](https://github.com/OpenBMB/MiniCPM-o) | [Demo](
|
20 |
|
21 |
|
22 |
|
23 |
## MiniCPM-V 4.0
|
24 |
|
25 |
-
**MiniCPM-V 4.0** is the latest model in the MiniCPM-V series. The model is built
|
26 |
|
27 |
- 🔥 **Leading Visual Capability.**
|
28 |
-
|
29 |
|
30 |
- 🚀 **Superior Efficiency.**
|
31 |
-
Designed for
|
32 |
|
33 |
- 💫 **Easy Usage.**
|
34 |
-
MiniCPM-V 4.0 can be easily used in various ways including **llama.cpp, Ollama, vLLM, SGLang, LLaMA-Factory and local web demo** etc. Get started easily with our
|
35 |
|
36 |
|
37 |
### Evaluation
|
|
|
16 |
|
17 |
<h1>A GPT-4V Level MLLM for Single Image, Multi Image and Video on Your Phone</h1>
|
18 |
|
19 |
+
[GitHub](https://github.com/OpenBMB/MiniCPM-o) | [Demo](http://211.93.21.133:8889/)</a>
|
20 |
|
21 |
|
22 |
|
23 |
## MiniCPM-V 4.0
|
24 |
|
25 |
+
**MiniCPM-V 4.0** is the latest efficient model in the MiniCPM-V series. The model is built based on SigLIP2-400M and MiniCPM4-3B with a total of 4.1B parameters. It inherits the strong single-image, multi-image and video understanding performance of MiniCPM-V 2.6 with largely improved efficiency. Notable features of MiniCPM-V 4.0 include:
|
26 |
|
27 |
- 🔥 **Leading Visual Capability.**
|
28 |
+
With only 4.1B parameters, MiniCPM-V 4.0 achieves an average score of 69.0 on OpenCompass, a comprehensive evaluation of 8 popular benchmarks, **outperforming GPT-4.1-mini-20250414, MiniCPM-V 2.6 (8.1B params, OpenCompass 65.2) and Qwen2.5-VL-3B-Instruct (3.8B params, OpenCompass 64.5)**. It also shows good performance in multi-image understanding and video understanding.
|
29 |
|
30 |
- 🚀 **Superior Efficiency.**
|
31 |
+
Designed for on-device deployment, MiniCPM-V 4.0 runs smoothly on end devices. For example, it devlivers **less than 2s first token delay and more than 17 token/s decoding on iPhone 16 Pro Max**, without heating problems. It also shows superior throughput under concurrent requests.
|
32 |
|
33 |
- 💫 **Easy Usage.**
|
34 |
+
MiniCPM-V 4.0 can be easily used in various ways including **llama.cpp, Ollama, vLLM, SGLang, LLaMA-Factory and local web demo** etc. We also open-source iOS App that can run on iPhone and iPad. Get started easily with our well-structured [Cookbook](https://github.com/OpenSQZ/MiniCPM-V-CookBook), featuring detailed instructions and practical examples.
|
35 |
|
36 |
|
37 |
### Evaluation
|