Multimodal Inference Optimization SGLang

MiniCPM-V 4.6

A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone

A 1.3B-parameter vision-language model release for efficient image, multi-image, video, and text understanding, optimized for edge and mobile deployment.

Pricing

Model profile · Current snapshot

Output TPS

TTFT (s)

Model Intelligence

Arena ranking

Benchmarkable

Yes

Model level

release

Intelligence Index

3.8

Coding Index

0.7

GPQA

0.31

HLE

0.05

SciCode

0.02

IFBench

0.27

LCR

0.07

TerminalBench Hard

TAU2

0.88

Recent stories

1 linked story

releasePRIMARY2026-05-11

OpenBMB releases MiniCPM-V 4.6 1.3B with 75.7 ms TTFT and 19x token efficiency

OpenBMB released MiniCPM-V 4.6 1.3B, claiming 55.8 percent lower vision-encoding FLOPs, 75.7 ms TTFT on a 4090, and about 1.5x token throughput over Qwen3.5 0.8B. It targets edge deployment across mobile platforms and common inference stacks.