6 models across 2 families · Latest: MOSS-Audio 4B Instruct (2026-04)
OpenMOSS audio and video foundation-model research.
MOSI Intelligence's portfolio covers 6 active models across 2 non-obsolete families, with task labels spanning vision. Open a model detail page to compare provider routes and sourced benchmarks.
Portfolio context: 1 decision-task tag, 6 active tracked models, latest research stamp 2026-06-04.
Use this portfolio page for
- Teams evaluating vision across this lab's releases
- Readers comparing families before locking a flagship SKU
- 6 tracked SKUs for migration and pricing follow-ups
Do not stop here for
- Choosing a hosting provider without opening a model page for price ladders
Active models
6
Non-deprecated SKUs linked to this researcher
Active families
2
Non-obsolete families in coverage
Open catalog
0 OSS
0 open-weight (text match)
Decision task tags
1
Mapped to the site-wide task taxonomy
Latest dated release
2026-04-13
MOSS-Audio 4B Instruct
Freshness
2026-06-04
Researched today
Release cadence
Showing 5 recent dated ships (full timeline below). Latest spotlight: MOSS-Audio 4B Instruct (2026-04-13).
Where this lab wins
- Vision: 6 tracked models with multimodal benchmark coverage.
Flagship quality / price signal
Anchor SKU: MOSS-Audio 4B Instruct (best sourced coding Q/$ in this portfolio).
Quality / dollar unavailable for this anchor — missing benchmark coverage and/or output token price on the cheapest ladder route (open the model detail after pricing lands).
MOSI Intelligence is a Chinese AI research organization. OpenMOSS audio and video foundation-model research. MOSI Intelligence ships 2 model families totaling 6 models, with the most recent release MOSS-Audio 4B Instruct in 2026-04. Notable families include MOSS-Audio and MOVA. Use it as a stable reference for lab background, release coverage, and follow-up model pages as they are added. Researchers and evaluators can. View official API endpoints, benchmark performance, and coding/agent fit for every MOSI Intelligence model.
About
MOSI Intelligence is the organization behind the OpenMOSS Team's open-weight audio and video foundation models, including MOSS-Audio for real-world audio understanding and MOVA for synchronized video-audio generation. Its OpenMOSS presence publishes research code, model cards, and weights through GitHub and Hugging Face, and should be tracked separately from Kyutai's Moshi voice model family.
Featured models
| Model | Released | Context | Input price ($/1M) | Output price ($/1M) | License |
|---|---|---|---|---|---|
| MOSS-Audio 4B Instruct | 2026-04-13 | - | - | - | Apache 2.0 |
| MOSS-Audio 4B Thinking | 2026-04-13 | - | - | - | Apache 2.0 |
| MOSS-Audio 8B Instruct | 2026-04-13 | - | - | - | Apache 2.0 |
Model families
Recent releases
- MOSS-Audio 4B Instruct- 2026-04-13
- MOSS-Audio 4B Thinking- 2026-04-13
- MOSS-Audio 8B Instruct- 2026-04-13
- MOSS-Audio 8B Thinking- 2026-04-13
- MOVA 360p- 2026-01-29
FAQ
What models has MOSI Intelligence released?
MOSI Intelligence ships 6 models across 2 families: MOSS-Audio and MOVA.
Is MOSI Intelligence's technology open source?
All tracked models are released under Apache 2.0.
Where is MOSI Intelligence headquartered?
MOSI Intelligence is headquartered in Shanghai, China.
What is MOSI Intelligence known for?
OpenMOSS audio and video foundation-model research. Its most prominent tracked family is MOSS-Audio.
How can I access MOSI Intelligence's models?
MOSI Intelligence's models are available via Hugging Face Inference Endpoints.
Explore related pages
Last reviewed: 2026-06-04. Data sourced from public lab announcements and provider documentation.