Quick start

ollama run minicpm-v4.6:latest

Runnable 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
1b	Typelocal	Inputstext, image	Digeste95583acac77	Download1.6 GB	Parameters1b	QuantizationUnknown	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:1b`
f16	Typelocal	Inputstext, image	Digest82520258dab3	Download2.6 GB	ParametersUnknown	QuantizationF16	Context256K	Est. memory≈ 4 GB	`ollama run minicpm-v4.6:f16`
latestdefault	Typelocal	Inputstext, image	Digeste95583acac77	Download1.6 GB	ParametersUnknown	QuantizationUnknown	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:latest`
q4_0	Typelocal	Inputstext, image	Digestfddef2599474	Download1.6 GB	ParametersUnknown	QuantizationQ4_0	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:q4_0`
q4_1	Typelocal	Inputstext, image	Digesta8ba7dee2228	Download1.6 GB	ParametersUnknown	QuantizationQ4_1	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:q4_1`
q4_K_M	Typelocal	Inputstext, image	Digeste95583acac77	Download1.6 GB	ParametersUnknown	QuantizationQ4_K_M	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:q4_K_M`
q4_K_S	Typelocal	Inputstext, image	Digest95200de9e025	Download1.6 GB	ParametersUnknown	QuantizationQ4_K_S	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:q4_K_S`
q5_0	Typelocal	Inputstext, image	Digest7d4df98d0fdc	Download1.7 GB	ParametersUnknown	QuantizationQ5_0	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:q5_0`
q5_1	Typelocal	Inputstext, image	Digest0132eed3d9ef	Download1.7 GB	ParametersUnknown	QuantizationQ5_1	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:q5_1`
q5_K_M	Typelocal	Inputstext, image	Digest0e195f4c8bb2	Download1.7 GB	ParametersUnknown	QuantizationQ5_K_M	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:q5_K_M`
q5_K_S	Typelocal	Inputstext, image	Digestb2b49d04e96a	Download1.7 GB	ParametersUnknown	QuantizationQ5_K_S	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:q5_K_S`
q6_K	Typelocal	Inputstext, image	Digestca9219b57000	Download1.7 GB	ParametersUnknown	QuantizationQ6_K	Context256K	Est. memory≈ 3 GB	`ollama run minicpm-v4.6:q6_K`
q8_0	Typelocal	Inputstext, image	Digestceb4ec05ddf6	Download1.9 GB	ParametersUnknown	QuantizationQ8_0	Context256K	Est. memory≈ 3.5 GB	`ollama run minicpm-v4.6:q8_0`

Related models

moondreamlocal

moondream2 is a small vision language model designed to run efficiently on edge devices.

glm-ocrlocal

GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture.

qwen3-vllocal

The most powerful vision-language model in the Qwen model family to date.

granite3.2-visionlocal

A compact and efficient vision-language model, specifically designed for visual document understanding, enabling automated content extraction from tables, charts, infographics, plots, diagrams, and more.