Kimi-Audio-7B
Open audio foundation model for understanding, generation, and speech conversation.
About
Moonshot AI's Kimi-Audio is a 7B open audio foundation model covering the full loop: speech recognition, audio understanding, audio-to-text chat, and speech-to-speech conversation, with a published technical report and a reproducible evaluation toolkit. Weights for both the base and instruct versions are on Hugging Face.
The license is a custom MIT-based license with additional terms, so check it before commercial use. Like all audio models, its conversation quality depends heavily on the eval setup — the published eval toolkit helps you verify claims yourself.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
- Strong results on audio understanding and generation evals (Moonshot-reported; eval toolkit published)
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (verified tier, metrics-v1). Evidence: Moonshot AI release; base weights 2025-04-25, instruct + tech report 2025-04-27; Published eval toolkit allows independent verification of claims; Custom MIT-based license with additional terms. License: Custom open license based on MIT with additional terms — check the repo LICENSE..
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 7.0/10
- Maintenance · recency, releases, issue load
- 7.0/10
- Openness · license + docs
- 7.0/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · Moonshot AI release; base weights 2025-04-25, instruct + tech report 2025-04-27
- · Published eval toolkit allows independent verification of claims
- · Custom MIT-based license with additional terms