Qwen3 Omni 30B A3B Instruct
Sep 2025
Qwen3 Omni 30B A3B InstructCompare inon
Qwen3 Omni 30B A3B Instruct measured on 3 of 34 Realtime benchmarks3
BenchmarkByScoreRank
Legacy · saturated or no longer reported1
No benchmark matches
self-host onlyprice
0.9 sfirst audio?
0.9Time to first audio
Sep 2025released
enLanguages
Alternatives?
ModelScoreCostPriceWhat changes
BetterGPT Live 1medium+59%—$0.05Languages: not verified+1
OpenStep-Audio R1.1+72%——Languages: not verified+1
BetterEleven Realtime v1 Max+74%—$0.08Languages: not verified+1
BetterGrok Voice 3.0+73%—$0.08Tool calling: not verifiedBetterGrok Voice 2.0unstated+71%—$0.08Languages: not verified+1ModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
GPT Live 1medium+59%—$0.05Languages: not verified+1Eleven Realtime v1 Max+74%—$0.08Languages: not verified+1Grok Voice 3.0+73%—$0.08Tool calling: not verifiedGrok Voice 2.0unstated+71%—$0.08Languages: not verified+1Eleven Realtime v1 Mini+34%—$0.08Languages: not verified+1ModelScoreCostPriceWhat changes
Step-Audio R1.1Open+72%——Languages: not verified+1
PersonaPlex 7BOpen-66%——Tool calling: not verifiedKMoshiOpen-92%——Tool calling: not verifiedModelScoreCostPriceWhat changes
Qwen Audio 3.0 Realtime Plus+75%——Languages: not verified+1Qwen 3.5 Omni Plus Realtime+74%——Languages: not verified+1Qwen Audio 3.0 Realtime Flash+69%——Languages: not verified+1Qwen 3.5 Omni Flash Realtime+4%——Languages: not verified+1Qwen 3.8 Omni Flash Realtimen/a——Languages: not verified+1Qwen Audio 3.1 Realtime Plusn/a——Languages: not verified+1Fun-Audio-Chat 8BOpenn/a——Tool calling: not verifiedQwen 3 Omni Flash Realtime 2025-09-15n/a——Languages: not verified+1All Alibaba (Qwen) models →Run 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
Benchmarks
in Speech & VoiceChat & Knowledge
Humanity's Last Exam4.6#290
AA Omniscience Index-69.3#292
CritPt (physics)0#180
AA Omniscience · Accuracy14.3#246
GPQA Diamond62#221
AIME 2025 · AA archived52.3#86
Coding
LiveCodeBench · AA archived42.2#101
Agents & Tool Use
τ²-bench · AA16.4#206
Writing & Language
IFBench (instructions)31.2#212
Vision & Documents
MMMU-Pro (vision)55.5#107
FAQ
How much does Qwen3 Omni 30B A3B Instruct cost?›
There is no per-unit API price: self-host only.
Is Qwen3 Omni 30B A3B Instruct free to use?›
There is no free API. The weights are open, so you can run it yourself.
How good is Qwen3 Omni 30B A3B Instruct, and what is it best at?›
#11 of 25 in Realtime.
Is Qwen3 Omni 30B A3B Instruct better than Raon SpeechChat 9B?›
On Big Bench Audio · AA v1.2, Qwen3 Omni 30B A3B Instruct scores 56.8 vs 57.5 — about the same.
When was Qwen3 Omni 30B A3B Instruct released? Is there a newer version?›
Released 2025-09-22. It is the current version in its line.
Is Qwen3 Omni 30B A3B Instruct open source?›
Yes, the weights are open (Apache-2.0 licence).
Can I run Qwen3 Omni 30B A3B Instruct locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Qwen3 Omni 30B A3B Instruct?›
Qwen3 Omni 30B A3B Instruct is an open-weights speech & voice model from Alibaba (Qwen), released Sep 2025.
What are the best alternatives to Qwen3 Omni 30B A3B Instruct?›
GPT Live 1 (better), Step-Audio R1.1 (open weights).