Open multimodal / omni generation models

Unified models that handle several modalities — image, audio, video and text — in one. Every one is downloadable and runs on your own machine or a rented GPU — honest, $0-markup.

Qwen2.5-Omni-7B — Omni understanding + speech generation, Alibaba Qwen · ~24 GB VRAM · rent $0.49/hr · commercial OK
Janus-Pro-7B — Unified understanding + image gen, DeepSeek · ~16 GB VRAM · rent $0.26/hr · commercial OK
OmniGen2 — Any to image (gen + edit + in context), BAAI / VectorSpaceLab · ~17 GB VRAM · rent $0.26/hr · commercial OK
MiniCPM-o 2.6 — Omni understanding + speech generation, OpenBMB · ~9.0 GB VRAM · rent $0.26/hr · commercial OK
BAGEL-7B-MoT — Unified understanding + image gen + edit, ByteDance Seed · ~24 GB VRAM · rent $0.49/hr · commercial OK
Emu3.5 — Any to any world model (gen + edit), BAAI · ~48 GB VRAM · rent $1.39/hr · commercial OK
Emu3-Gen — Next token any to any generation, BAAI · ~18 GB VRAM · rent $0.26/hr · commercial OK
OmniGen v1 — Unified image gen + edit, BAAI · ~12 GB VRAM · rent $0.26/hr · commercial OK
Ming-Lite-Omni — Omni understanding + image & speech gen, inclusionAI (Ant Group) · ~41 GB VRAM · rent $0.49/hr · commercial OK
Janus-Pro-1B — Unified understanding + image gen, DeepSeek · ~6.0 GB VRAM · rent $0.26/hr · commercial OK
Show-o2-7B — Unified understanding + image/video gen, Show Lab (NUS) · ~18 GB VRAM · rent $0.26/hr · commercial OK
Lumina-mGPT 2.0 — Autoregressive image gen + edit, Alpha-VLLM (Shanghai AI Lab) · ~24 GB VRAM · rent $0.49/hr · commercial OK

Other media types

Image · Video · Voice & Audio · All media