Quick start
ollama run minicpm-v4.5:latestRunnable variants
Memory is a rough file-size estimate; context length, runtime and GPU offload change it.
12 of 13 matching variants shown.13 of 13 variants shown.
| Tag | Type | Inputs | Digest | Download | Parameters | Quantization | Context | Est. memory | Run |
|---|---|---|---|---|---|---|---|---|---|
| 8b | Typelocal | Inputstext, image | Digest0c40168f46d1 | Download6.1 GB | Parameters8b | QuantizationUnknown | Context40K | Est. memory≈ 8 GB | ollama run minicpm-v4.5:8b |
| f16 | Typelocal | Inputstext, image | Digeste8c96bb677b7 | Download17 GB | ParametersUnknown | QuantizationF16 | Context40K | Est. memory≈ 20 GB | ollama run minicpm-v4.5:f16 |
| latestdefault | Typelocal | Inputstext, image | Digest0c40168f46d1 | Download6.1 GB | ParametersUnknown | QuantizationUnknown | Context40K | Est. memory≈ 8 GB | ollama run minicpm-v4.5:latest |
| q4_0 | Typelocal | Inputstext, image | Digestfd2ea418f30e | Download5.9 GB | ParametersUnknown | QuantizationQ4_0 | Context40K | Est. memory≈ 8 GB | ollama run minicpm-v4.5:q4_0 |
| q4_1 | Typelocal | Inputstext, image | Digeste48898b0d902 | Download6.3 GB | ParametersUnknown | QuantizationQ4_1 | Context40K | Est. memory≈ 8.5 GB | ollama run minicpm-v4.5:q4_1 |
| q4_K_M | Typelocal | Inputstext, image | Digest0c40168f46d1 | Download6.1 GB | ParametersUnknown | QuantizationQ4_K_M | Context40K | Est. memory≈ 8 GB | ollama run minicpm-v4.5:q4_K_M |
| q4_K_S | Typelocal | Inputstext, image | Digest8a85b33d3053 | Download5.9 GB | ParametersUnknown | QuantizationQ4_K_S | Context40K | Est. memory≈ 8 GB | ollama run minicpm-v4.5:q4_K_S |
| q5_0 | Typelocal | Inputstext, image | Digest2bc93de9ac8e | Download6.8 GB | ParametersUnknown | QuantizationQ5_0 | Context40K | Est. memory≈ 9 GB | ollama run minicpm-v4.5:q5_0 |
| q5_1 | Typelocal | Inputstext, image | Digest455fdec4b9e8 | Download7.3 GB | ParametersUnknown | QuantizationQ5_1 | Context40K | Est. memory≈ 9.5 GB | ollama run minicpm-v4.5:q5_1 |
| q5_K_M | Typelocal | Inputstext, image | Digestbea3ec44177e | Download6.9 GB | ParametersUnknown | QuantizationQ5_K_M | Context40K | Est. memory≈ 9 GB | ollama run minicpm-v4.5:q5_K_M |
| q5_K_S | Typelocal | Inputstext, image | Digestc714a3479462 | Download6.8 GB | ParametersUnknown | QuantizationQ5_K_S | Context40K | Est. memory≈ 9 GB | ollama run minicpm-v4.5:q5_K_S |
| q6_K | Typelocal | Inputstext, image | Digest6ca2c521c157 | Download7.8 GB | ParametersUnknown | QuantizationQ6_K | Context40K | Est. memory≈ 10 GB | ollama run minicpm-v4.5:q6_K |
| q8_0 | Typelocal | Inputstext, image | Digestd3a452c30a41 | Download9.8 GB | ParametersUnknown | QuantizationQ8_0 | Context40K | Est. memory≈ 12 GB | ollama run minicpm-v4.5:q8_0 |
Related models
llava-llama3local
A LLaVA model fine-tuned from Llama 3 Instruct with better scores in several benchmarks.
ornith-1.5local
Chirp Chirp! 🐦 We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement.
deepseek-ocrlocal
DeepSeek-OCR is a vision-language model that can perform token-efficient OCR.
gemma3nlocal
Gemma 3n models are designed for efficient execution on everyday devices such as laptops, tablets or phones.