MiniCPM-V-4_6

Checkpointopenbmb/MiniCPM-V-4_6Model card indexedModel card

Support summary

MiniCPM-V-4_6 is a checkpoint published by OpenBMB, built on the MiniCPMV and MiniCPMV4_6ForConditionalGeneration architectures. vLLM 0.28.0, the newest indexed vLLM release, documents support. Official recipes or cookbooks cover SGLang, without pinning a stable release.

vLLM documents support in 6 of 20 indexed releases, from 0.25.0 to 0.28.0.

Documented quantization
BF16
Configured context
262,144 tokens
License
apache-2.0

Hardware, quantization and features count only statements about this checkpoint; engine-wide capabilities are listed with each release.

Hugging Face model card
Publisher
OpenBMB
Task
image text to text
License
apache-2.0
Library
transformers
Architecture
MiniCPMV4_6ForConditionalGeneration
Weight tensor elements
1.3B
Tensor types
BF16
Configured context
262,144 tokens
Weight access
Public
Repository created
2026-04-13
Source 1: openbmb/MiniCPM-V-4.6 · publisher model cardSource 2: openbmb/MiniCPM-V-4.6 · Hugging Face metadataSource 3: openbmb/MiniCPM-V-4.6 · publisher configurationReviewed
Model notes and sources

Engine support

3 engines

Engine support for MiniCPM-V-4_6
EngineReleaseCheckpoint supportEvidence scopeLast known supported
vLLM0.28.0Newest indexedCheckpoint evidenceSource: vLLM v0.28.0 · docs/models/supported_models.mdReviewed 0.28.0This release
Features, dependencies and evidence for vLLM 0.28.0
SGLangVersion not pinnedRecipe / cookbookDocumented recipe

Official recipe

Source 1: sglang · docs/cookbook/autoregressive/OpenBMB/MiniCPM-V-4_6.mdxSource 2: sglang · docs/src/snippets/autoregressive/minicpm-v-4_6-deployment.jsxReviewed
No supported release indexed
Features, dependencies and evidence for SGLang 0.5.19
TensorRT-LLM1.2.1Newest indexedNo matching evidence curatedNo supported release indexed
Features, dependencies and evidence for TensorRT-LLM 1.2.1

Recipes and cookbooks