Chat
Qwen3 30B A3B
Qwen3 30B A3B is the mainstream sparse mixture-of-experts Qwen - 30.5B total parameters, 128 experts, roughly 3B active per token - and the current Ollama tag serves the 2507 refresh with a 256K context window.
Alibaba Qwen · Qwen
Editorial review
Model checkpoints, context windows, provider support, local runtime compatibility, and license terms can change quickly. Verify the exact model card before production or commercial use.
Best for
High-quality local chat, reasoning, and agent workflows on mid-range GPUs - sparsity keeps split-mode performance usable.
Who should use it
- High-quality local chat, reasoning, and agent workflows on mid-range GPUs - sparsity keeps split-mode performance usable.
- Builders who want local or self-hosted testing options.
Common workflows
- Multilingual chat, reasoning, and assistant workflows
- chat workflows
- multilingual workflows
- reasoning workflows
- open weights workflows
Deployment and hardware notes
Ollama Q4_K_M is ~19 GB: full-GPU on 24 GB cards, usable split placement on 8-12 GB GPUs with fast system RAM thanks to ~3B active parameters per token.
License and usage notes
Apache 2.0. Open weights. Verify the exact model card and license terms for the checkpoint or hosted provider you use.
Strengths
- Open weights model option for Qwen workflows.
- High-quality local chat, reasoning, and agent workflows on mid-range GPUs - sparsity keeps split-mode performance usable.
- Use the exact checkpoint and quantization that matches your hardware and latency target.
- Tracked as Frontier 2026 in the OpenSourcesAI model directory.
Limitations
- Q4 weights are ~19 GB, so full-GPU placement needs a 24 GB card; measured 31.6 tok/s on an RTX 3080 (10 GB) at 46% VRAM residency (2026-08-01), so split mode remains genuinely usable. Verified against the Ollama tag and HF config.
- Ollama Q4_K_M is ~19 GB: full-GPU on 24 GB cards, usable split placement on 8-12 GB GPUs with fast system RAM thanks to ~3B active parameters per token.
- Context window and limits: 256K tokens (tag-served 2507 refresh; verified via ollama show 2026-08-01).
- Verify the exact model card, provider docs, license, and serving support before production use.
Local workflow notes
Use the exact checkpoint and quantization that matches your hardware and latency target.
Local runtimes: Ollama where supported, LM Studio where supported, llama.cpp where supported, Transformers
Platforms: Windows, macOS, Linux
Will Qwen3 30B (A3B) run on your machine?
Qwen3 30B (A3B) is 30.5B parameters and needs 20.5 GB of VRAM at Q4_K_M — 19 GB of weights plus 1.5 GB of runtime overhead for the inference server itself.
VRAM by quantization
| Quantization | Weights | Needs (with overhead) | Quality |
|---|---|---|---|
| Q4_K_M | 19 GB | 20.5 GB | good |
Fit on common hardware at Q4_K_M
| Hardware | Memory the model can use | System RAM | Verdict |
|---|---|---|---|
| CPU Only | None (CPU only) | 16 GB | Too large |
| RTX 4060 Laptop | 8 GB | 16 GB | Too large |
| RTX 3060 (12GB) | 12 GB | 32 GB | CPU offload |
| RTX 4060 Ti (16GB) | 16 GB | 32 GB | CPU offload |
| RTX 3090 | 24 GB | 64 GB | Comfortable |
| Apple Silicon (Unified Memory) 36 GB | 27 GB of 36 GB | 36 GB | Comfortable |
| RTX 5090 | 32 GB | 64 GB | Comfortable |
Comfortable means VRAM clears the requirement by 2 GB or more. Tight means it covers the requirement with no margin. CPU offload means the model does not fit in VRAM but system RAM is at least 1.6× the weights, so it will run at reduced speed — expect roughly 1–5 tokens per second. Figures are weights plus a fixed runtime overhead and exclude KV-cache growth, which scales with context length.
Apple Silicon shares one pool of memory between the system and the GPU, so a model cannot use all of it. These rows apply the same 75% usable fraction the Compatibility Checker uses, which is why a 36 GB Mac is graded on less than 36 GB.
VRAM fit by quantization level
Enter your GPU VRAM below to see which quantization of Qwen3 30B A3B fits and get the Ollama run command.
Frontier-model verification note
This page is written to stay accurate as of the latest available 2026 public model information. Availability, licenses, context windows, API support, pricing, benchmark standing, and local-serving support can change quickly. Verify the official model card, provider docs, and license before using this model in production or commercial workflows.
Sources to verify
Related resources
Continue with model source notes, local tools, and implementation guides related to this model.
Model ecosystem connections
Use these next-step links to move from this profile into related tools, comparisons, guides, stacks, and curated shortlists.
Recommended runtimes and tools
Setup and deployment
Related model pages
Guides, stacks, and comparisons
Ready to run this model locally?
Find a compatible interface in our Local AI Tools directory →