Step-Audio R1.1
Jan 2026
Step-Audio R1.1Compare inon
Step-Audio R1.1 measured on 2 of 34 Realtime benchmarks2
BenchmarkByScoreRank
Legacy · saturated or no longer reported1
No benchmark matches
self-host onlyprice
1.5 sfirst audio?
1.5Time to first audio
Jan 2026released
Alternatives?
ModelScoreCostPriceWhat changes
OpenQwen3 Omni 30B A3B Instructunstatedsame——Memory (4-bit): 20 GB → 21 GBSame labStepAudio 3 Realtime Preview-83%——ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
Qwen3 Omni 30B A3B InstructunstatedOpensame——Memory (4-bit): 20 GB → 21 GBRun it yourself
estimateFits on✗ 8 GB (phone, laptop)✗ 16 GB✓ 24 GB GPU · compressed✓ 48 GB GPU · 8-bit✓ 80 GB GPU · full quality
LicenceApache-2.0free for commercial use
FAQ
How much does Step-Audio R1.1 cost?›
There is no per-unit API price: self-host only.
Is Step-Audio R1.1 free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Step-Audio R1.1, and what is it best at?›
#21 of 25 in Realtime.
Is Step-Audio R1.1 better than Higgs Realtime?›
On Time to first audio · AA, Step-Audio R1.1 scores 1.5 vs 1.5 — about the same.
When was Step-Audio R1.1 released? Is there a newer version?›
Released 2026-01-14. It is the current version in its line.
Is Step-Audio R1.1 open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Step-Audio R1.1 locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Step-Audio R1.1?›
Step-Audio R1.1 is an open-weights speech & voice model from StepFun, released Jan 2026.
What are the best alternatives to Step-Audio R1.1?›
Qwen3 Omni 30B A3B Instruct (open weights).