Run Qwen3-30B-A3B on-premises
○ Memory-fit arithmetic — checkable math, not a benchmark
Qwen3-30B-A3B is a mixture-of-experts model with 30 billion total parameters and about 3 billion active per token, under the permissive Apache-2.0 license. Its small active footprint makes it fast and cheap to serve, and it leaves so much memory free that one machine can host it alongside other models.
Here is the arithmetic against a machine with 128 GB of memory.
Memory fit in 128 GB
| Parameters | 30B total, ~3B active per token (MoE) |
|---|---|
| 4-bit weights | ~15–18 GB |
| Reference memory | 128 GB (one machine) |
| Left for context + other models | roughly 100 GB |
| Fits in 128 GB | yes, with room for more models |
At ~15–18 GB the model uses a small fraction of 128 GB, so the same machine can serve it plus an embedding model and a coder model concurrently.
What it serves well
- High-throughput chat, routing, extraction, and agent steps at low serving cost.
- Multi-model machines where several small/mid models run side by side.
- Latency-sensitive workloads that don't need a 100B-class brain.
Honest limits
- Memory arithmetic, not a benchmark — and here the math is straightforward and generous.
- A ~3B active budget is efficient, not maximal — for the hardest reasoning we would pair or route to a bigger model, which the same 128 GB can also hold.
Frequently asked questions
Can one machine run this plus other models?
Yes — at ~15–18 GB it leaves roughly 100 GB of 128 GB free, enough to co-host embedding and coder models on the same machine.
License?
Apache-2.0 — permissive and commercial-friendly.
Is it fast?
Its ~3B active-parameter budget makes it markedly faster to serve than a dense model of similar total size; we quantify it for your workload in the assessment.
Find out which model fits your work
An AI diagnostic maps your processes, classes what is sensitive and sizes the model and the hardware against your real tasks.