MiniCPM-V — lokalny model AI od OpenBMB
Autor: Jakub Rusinowski · Ostatnia aktualizacja:
OpenBMB's pocket-sized multimodal line, backed by ModelBest and Tsinghua University, built around one goal: image and video understanding that actually runs on a phone. Version 4.5 is the 8B that beat much larger proprietary models on OpenCompass; version 4.6 is the 1.3B successor that cut visual-encoding FLOPs by more than half and ships on iOS, Android and HarmonyOS.
Warianty
Najmniejszy wariant MiniCPM-V potrzebuje około 2 GB VRAM przy Q4_K_M — skwantyzowane wagi plus narzut frameworka, bez pamięci podręcznej KV.
| Model | VRAM przy Q4 | VRAM | Kontekst | Uruchom |
|---|---|---|---|---|
| MiniCPM-V 4.6 → 1.3B | ~1.6 GB | 262,144 | ollama run minicpm-v4.6 | |
| MiniCPM-V 4.5 → 8B | ~5.6 GB | 32,768 | ollama run minicpm-v4.5 |
Pamięć to skwantyzowane wagi plus narzut przy Q4_K_M, liczone tym samym silnikiem co sprawdzarka GPU i VRAM.
Jak uruchomić MiniCPM-V lokalnie
Zainstaluj Ollama, a potem pobierz tag.
ollama run minicpm-v4.6Wybierz rozmiar powyżej, aby zobaczyć jego VRAM, szacowaną szybkość i polecenie instalacji.
Licencja
Commercial use permitted. No usage restrictions beyond attribution.
Dotyczy: MiniCPM-V 4.6Weights are downloadable and commercial use is permitted, subject to the licence’s acceptable-use terms.
Dotyczy: MiniCPM-V 4.5Polecane GPU
Najtańsze GPU z naszego katalogu, które uruchomi MiniCPM-V lokalnie (min. 2 GB VRAM), to Intel Arc B570 (10 GB).
Czy uruchomię MiniCPM-V na moim GPU?
- MiniCPM-V na AMD Radeon RX 9060 XT 8GB
- MiniCPM-V na Intel Arc B570
- MiniCPM-V na Intel Arc B580
- MiniCPM-V na NVIDIA GeForce RTX 3060 (12GB)
- MiniCPM-V na NVIDIA GeForce RTX 3070
- MiniCPM-V na NVIDIA GeForce RTX 3070 Ti
- MiniCPM-V na NVIDIA GeForce RTX 3080 (10GB)
- MiniCPM-V na NVIDIA GeForce RTX 4060
- MiniCPM-V na NVIDIA GeForce RTX 4070
- MiniCPM-V na NVIDIA GeForce RTX 4070 Super
- MiniCPM-V na NVIDIA GeForce RTX 4070 Ti
- MiniCPM-V na NVIDIA GeForce RTX 5060
- MiniCPM-V na NVIDIA GeForce RTX 5060 Ti 8GB
- MiniCPM-V na NVIDIA GeForce RTX 5070
MiniCPM-V — najczęściej zadawane pytania
How much VRAM does MiniCPM-V need?
MiniCPM-V needs about 2 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: MiniCPM-V 4.6 (2 GB, Q4_K_M); MiniCPM-V 4.5 (6 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run MiniCPM-V on an RTX 4090 (24 GB)?
Yes — MiniCPM-V runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
What quantization should I use for MiniCPM-V?
Q4_K_M is the best balance of quality and VRAM for MiniCPM-V in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run MiniCPM-V with Ollama?
Install Ollama, then run: ollama run minicpm-v4.6. This downloads MiniCPM-V and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.