Qwen 3 32B needs ~26.7 GB VRAM. Intel Arc Pro B60 24GB has 24.0 GB. With Q4_K_M quantization, expect ~8 tok/s.
Operating mode
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
2.7 GB over capacity — needs offload or smaller quantization
Fit status
Very compromised (needs ~2 GB host RAM)
Decode
8.4 tok/s
TTFT
22984 ms
Safe context
5K
Memory
26.7 GB / 24.0 GB
Offload
10%
It fits through host-memory offload, and offload is the main reason performance drops.
CPU or host-memory offload is active
About 10% of the working set spills out of accelerator memory, which usually hurts latency and sustained decode throughput.
Very little memory headroom
You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.
Runtime ecosystem is narrower than CUDA
Intel GPUs can look attractive on memory per dollar, but local AI tooling, kernels, and model coverage are still broader and easier on CUDA today.
Remove offload with more accelerator memory
Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.
Prefer CUDA if you want the path of least resistance
If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.
Buy headroom, not only minimum fit
A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | S | Runs with offload (needs ~0.6 GB host RAM) | 9.8 tok/s | 10739 ms | 5K |
| Coding | A | Very compromised (needs ~2 GB host RAM) | 8.4 tok/s | 22984 ms | 5K |
| Agentic Coding | F | Too heavy | 6.4 tok/s | 44158 ms | 5K |
| Reasoning | A | Very compromised (needs ~2 GB host RAM) | 8.4 tok/s | 27163 ms | 5K |
| RAG | F | Too heavy | 6.4 tok/s | 55198 ms |
How Qwen 3 32B (32B params) fits at each quantization level on Intel Arc Pro B60 24GB (24.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 12.5 GB | Low | S91 |
Q3_K_S | 3 | 15.7 GB | Low | S91 |
NVFP4Best for your GPU |
Copy-paste commands to run Qwen 3 32B on your machine.
Run
ollama run qwen3:32bYour hardware
| Model | Params | Grade | Decode | Capabilities |
|---|---|---|---|---|
 Qwen 3.6 35B A3B | 35B | A | 16.6 tok/s | |
|
Yes, Intel Arc Pro B60 24GB can run Qwen 3 32B with a A grade (Very compromised (needs ~2 GB host RAM)). Expected decode speed: 8.4 tok/s.
Qwen 3 32B (32B parameters) requires approximately 26.7 GB of memory with Q4_K_M quantization.
The recommended quantization for Qwen 3 32B is Q4_K_M, which balances quality and memory efficiency.
On Intel Arc Pro B60 24GB, Qwen 3 32B achieves approximately 8.4 tokens per second decode speed with a time-to-first-token of 22984ms using Q4_K_M quantization.
For coding workloads, Qwen 3 32B on Intel Arc Pro B60 24GB receives a A grade with 8.4 tok/s and 5K context.
On Intel Arc Pro B60 24GB, Qwen 3 32B can safely use up to 5K tokens of context. The model's official context limit is 131K, but available memory constrains the safe maximum.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/qwen-3-32b-on-arc-pro-b60-24gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview:
| 5K |
| 4 |
17.9 GB |
| Medium |
| S90 |
Q4_K_M | 4 | 19.5 GB | Medium | F0 |
Q5_K_M | 5 | 23.0 GB | High | F0 |
Q6_K | 6 | 26.2 GB | High | F0 |
Q8_0 | 8 | 34.2 GB | Very High | F0 |
F16 | 16 | 65.6 GB | Maximum | F0 |
| 35B |
| A |
| 21.9 tok/s |
Remove offload with more accelerator memory. Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.
Often yes, if your goal is the easiest setup and the widest runtime support. Intel can offer attractive memory capacity, but CUDA still tends to win on tooling maturity, guides, kernels, and model coverage for local AI.