FitMyLLM
Alibaba/Mixture of Experts

AlibabaQwen3-Omni 30B-A3B

Truly multimodal: text, images, audio, video input. MoE 30B (3B active). 128K context. 119 languages.

chatvisionmultilingual
30B
Parameters (3B active)
128K
Context length
19
Benchmarks
14
Quantizations
Architecture
MoE
Released
2026-02-15
Layers
36
KV Heads
4
Head Dim
128
Family
qwen

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
IQ3_XXS3.25
13.5 GB
12.7 + 0.8 KV
low
IQ3_XS3.5
14.5 GB
13.6 + 0.8 KV
low
Q3_K_S3.64
15.0 GB
14.1 + 0.8 KV
low
IQ3_M3.76
15.4 GB
14.6 + 0.8 KV
low
Q3_K_M4
16.3 GB
15.5 + 0.8 KV
low
Q3_K_L4.3
17.5 GB
16.6 + 0.8 KV
moderate
IQ4_XS4.46
18.1 GB
17.2 + 0.8 KV
moderate
Q4_K_S4.67
18.8 GB
18.0 + 0.8 KV
moderate
Q4_K_M4.89
19.7 GB
18.8 + 0.8 KV
good
Q5_K_S5.57
22.2 GB
21.4 + 0.8 KV
good
Q5_K_M5.7
22.7 GB
21.9 + 0.8 KV
good
Q6_K6.56
25.9 GB
25.1 + 0.8 KV
excellent
Q8_08.5
33.2 GB
32.4 + 0.8 KV
lossless
FP1616
61.3 GB
60.5 + 0.8 KV
lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Deploying for a team or in production? Size GPUs, cost & scaling in Enterprise →
READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN QWEN3-OMNI 30B-A3B NOW

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (19)

IFEval78.9
AIME74.0
AA Math74.0
LiveCodeBench67.9
MMLU-PRO61.5
MATH59.7
BBH58.3
MATH-50052.3
GPQA Diamond43.9
IFBench43.4
BigCodeBench32.3
SciCode30.6
τ²-Bench21.3
MUSR19.1
AA Intelligence15.6
GPQA15.2
AA Coding12.7
HLE7.3
Terminal-Bench3.8

Run this model

Easiest way to get started·Beginners
DOCS ↗
curl -fsSL https://ollama.com/install.sh | sh
$ollama run qwen:30b-q4_K_M

Tag may need adjustment — check ollama.com/library/qwen for available tags.

▸ SETUP GUIDE
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

NVIDIA RTX A4500
20 GB VRAM • 640 GB/s
NVIDIA
$2000
Apple M4 Pro (24GB)
24 GB VRAM • 273 GB/s
APPLE
$1399
NVIDIA L4 24GB
24 GB VRAM • 300 GB/s
NVIDIA
$2500
Apple M2 (24GB)
24 GB VRAM • 100 GB/s
APPLE
$999
Apple M3 (24GB)
24 GB VRAM • 100 GB/s
APPLE
$999
Apple M4 (24GB)
24 GB VRAM • 120 GB/s
APPLE
$699
NVIDIA Tesla M40 24 GB
24 GB VRAM • 288 GB/s
NVIDIA
NVIDIA Tesla P10
24 GB VRAM • 694 GB/s
NVIDIA
NVIDIA Tesla P40
24 GB VRAM • 347 GB/s
NVIDIA
NVIDIA RTX A5000
24 GB VRAM • 768 GB/s
NVIDIA
$2500
NVIDIA L40 CNX
24 GB VRAM • 864 GB/s
NVIDIA
$5000

Find the best GPU for Qwen3-Omni 30B-A3B

Build Hardware for Qwen3-Omni 30B-A3B
▸ SPEC SHEET

Qwen3-Omni 30B-A3B30B MoE.

▸ SPECIFICATIONS
PARAMETERS
30B (3B active)
ARCHITECTURE
Mixture of Experts
CONTEXT LENGTH
128K tokens
CAPABILITIES
chat, vision, multilingual
RELEASE DATE
2026-02-15
PROVIDER
Alibaba
FAMILY
qwen
▸ VRAM REQUIREMENTS
QUANTBPWVRAMQUALITY
IQ3_XXS3.2512.7 GB82%
IQ3_XS3.513.6 GB84%
Q3_K_S3.6414.1 GB85%
IQ3_M3.7614.6 GB86%
Q3_K_M415.5 GB88%
Q3_K_L4.316.6 GB90%
IQ4_XS4.4617.2 GB92%
Q4_K_S4.6718.0 GB93%
Q4_K_M4.8918.8 GB94%
Q5_K_S5.5721.4 GB96%
Q5_K_M5.721.9 GB96%
Q6_K6.5625.1 GB97%
Q8_08.532.4 GB100%
FP161660.5 GB100%
§ 01BENCHMARK SCORES
MMLU-PRO61.5
MATH59.7
IFEval78.9
BBH58.3
GPQA15.2
MUSR19.1
BigCodeBench32.3
LiveCodeBench67.9
AIME74.0
MATH-50052.3
GPQA Diamond43.9
HLE7.3
AA Intelligence15.6
AA Coding12.7
AA Math74.0
aa_ifbench43.4
aa_terminal_bench3.8
aa_tau221.3
aa_scicode30.6
§ 02RUN COMMAND

Run Qwen3-Omni 30B-A3B locally with Ollama — needs 18.8 GB VRAM at Q4_K_M:

$ollama run qwen3-omni:30b
§ 03COMPATIBLE GPUs
30 @ Q4_K_M