selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

empero-ai logo

Can the Arc B580 run Qwen3.8-4B-Distill?

Qwen3.8-4B-Distill·4.7B·Arc B580·12 GB VRAM

Saved in this browser, used on every model page.
Yes

Yes. The Arc B580 runs Qwen3.8-4B-Distill fully in VRAM at FP16 / BF16 (about 10.0 GB of 12 GB), at roughly 27 tokens/s.

best quant
FP16 / BF16
needs
10.0 GB
usable
12 GB
est. speed
~27 tok/s

Estimates for an 8K context; offload assumes 32 GB of system RAM.Try other contexts on the Qwen3.8-4B-Distill page →

Every quantisation of Qwen3.8-4B-Distill on the Arc B580

QuantQualityMemoryFitSpeed
Q2_K Heavy loss3.4 GBfits~164 tok/s
Q3_K_M Noticeable loss3.8 GBfits~110 tok/s
Q4_K_M Small loss4.2 GBfits~89 tok/s
Q5_K_M Near-lossless4.6 GBfits~76 tok/s
Q6_K Near-lossless5.0 GBfits~65 tok/s
Q8_0 Lossless6.0 GBfits~51 tok/s
FP16 / BF16 pickLossless10.0 GBfits~27 tok/s

Speed is a bandwidth-based estimate for generating text (456 GB/s); prompt processing and long contexts are slower.

Measured speed on the Arc B580

No community reports for this pair yet.

Report your own speed →

Run Qwen3.8-4B-Distill on the Arc B580

Using FP16 / BF16, the best fit for Arc B580.

Ollama · BF16 straight from Hugging Facedocs
ollama run hf.co/empero-ai/Qwen3.8-4B-Distill-GGUF:BF16

Ollama’s ROCm/CUDA builds don’t cover this GPU. Recent Ollama releases use Vulkan when its Vulkan backend and driver are installed (older ones needed OLLAMA_VULKAN=1); llama.cpp or LM Studio with Vulkan are the safer bet.

llama.cpp · OpenAI-compatible server with a web UIdocs
llama-server -hf empero-ai/Qwen3.8-4B-Distill-GGUF:BF16

Needs a Vulkan build of llama.cpp. Downloads the GGUF on first run.

llama.cpp · Docker with Vulkandocs
docker run -d --name llama --device /dev/dri -p 8080:8080 \
  -v ~/.cache/llama.cpp:/root/.cache/llama.cpp \
  ghcr.io/ggml-org/llama.cpp:server-vulkan \
  -hf empero-ai/Qwen3.8-4B-Distill-GGUF:BF16 --host 0.0.0.0 --port 8080 -ngl 99

Server and web UI on port 8080.

LM Studio · Download with the lms CLIdocs
lms get empero-ai/Qwen3.8-4B-Distill-GGUF@BF16

Then load it from the LM Studio app or with `lms load`.

Transformers · Chat with the pipeline APIdocs
from transformers import pipeline

pipe = pipeline("text-generation", model="empero-ai/Qwen3.8-4B-Distill", dtype="auto", device_map="auto")
messages = [{"role": "user", "content": "Explain self-hosting in one paragraph."}]
out = pipe(messages, max_new_tokens=512)
print(out[0]["generated_text"][-1]["content"])

Chat models that fit the Arc B580 comfortably

All that fit →
ModelParamsBest quantMemorySpeedLicence
4BQ8_0≈5.8 GBFits in memory~59 tok/sCommercial OK
3.2BQ8_0≈4.7 GBFits in memory~73 tok/sConditions
7.6BQ8_0≈8.9 GBFits in memory~31 tok/sCommercial OK
8BQ8_0≈9.8 GBFits in memory~29 tok/sConditions
3.8BQ8_0≈5.5 GBFits in memory~62 tok/sCommercial OK
4BQ8_0≈5.8 GBFits in memory~59 tok/sCommercial OK

Other GPUs that run Qwen3.8-4B-Distill

Frequently asked questions

Can the Arc B580 run Qwen3.8-4B-Distill?
Yes. The Arc B580 runs Qwen3.8-4B-Distill fully in VRAM at FP16 / BF16 (about 10.0 GB of 12 GB), at roughly 27 tokens/s.
How fast does Qwen3.8-4B-Distill run on the Arc B580?
An estimate from memory bandwidth gives about 27 tokens/s at FP16 / BF16.
Which quantisation of Qwen3.8-4B-Distill should I use on the Arc B580?
FP16 / BF16: the largest one that fits. Expected quality: lossless compared with the full-precision weights.

Memory and speed are estimates from the model's size and the card's memory bandwidth; real numbers depend on the runtime, drivers and context length. Community figures are medians of user reports.