Can the Arc A770 16GB run ZwZ2-27B?
ZwZ2-27B·27.4B·Arc A770 16GB·16 GB VRAM
Just about. At Q3_K_M it needs about 16 GB of the 16 GB available, so keep the context short, at roughly 23 tokens/s.
- best quant
- Q3_K_M
- needs
- 16 GB
- usable
- 16 GB
- est. speed
- ~23 tok/s
Estimates for an 8K context; offload assumes 32 GB of system RAM.Try other contexts on the ZwZ2-27B page →
Every quantisation of ZwZ2-27B on the Arc A770 16GB
| Quant | Memory | Fit | Speed |
|---|---|---|---|
| Q2_K | 11 GB | fits | ~34 tok/s |
| Q3_K_M pick | 16 GB | tight | ~23 tok/s |
| Q4_K_M | 19 GB | offload | ~8.5 tok/s |
| Q5_K_M | 21 GB | offload | ~5.1 tok/s |
| Q6_K | 24 GB | offload | ~3.6 tok/s |
| Q8_0 | 30 GB | offload | ~2.1 tok/s |
| FP16 / BF16 | 55 GB | too big | — |
Speed is a bandwidth-based estimate for generating text (560 GB/s); prompt processing and long contexts are slower.
Measured speed on the Arc A770 16GB
No community reports for this pair yet.
Report your own speed →Run ZwZ2-27B on the Arc A770 16GB
Using Q3_K_M, the best fit for Arc A770 16GB.
llama-server -hf prithivMLmods/inclusionAI-ZwZ2-27B-GGUF:Q3_K_LNeeds a Vulkan build of llama.cpp. Downloads the GGUF on first run.
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 prithivMLmods/inclusionAI-ZwZ2-27B-GGUF:Q3_K_L --host 0.0.0.0 --port 8080 -ngl 99Server and web UI on port 8080.
from transformers import pipeline
pipe = pipeline("image-text-to-text", model="inclusionAI/ZwZ2-27B", dtype="auto", device_map="auto")
messages = [
{
"role": "user",
"content": [
{"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},
{"type": "text", "text": "Describe this image."},
],
}
]
out = pipe(text=messages, max_new_tokens=256)
print(out[0]["generated_text"][-1]["content"])Vision models that fit the Arc A770 16GB comfortably
All that fit →| Model | Params | Best quant | Memory | Speed |
|---|---|---|---|---|
| 12.2B | Q6_K | ≈13 GBFits in memory | ~31 tok/s | |
| 19.4BA3B | Q4_K_M | ≈13 GBFits in memory | ~169 tok/s | |
| 9.7B | Q8_0 | ≈11 GBFits in memory | ~30 tok/s | |
| 13.2B | Q6_K | ≈13 GBFits in memory | ~28 tok/s | |
| 8.8B | Q8_0 | ≈10 GBFits in memory | ~33 tok/s | |
| 10.7B | Q8_0 | ≈12 GBFits in memory | ~27 tok/s |
Other GPUs that run ZwZ2-27B
Frequently asked questions
- Can the Arc A770 16GB run ZwZ2-27B?
- Just about. At Q3_K_M it needs about 16 GB of the 16 GB available, so keep the context short, at roughly 23 tokens/s.
- How fast does ZwZ2-27B run on the Arc A770 16GB?
- An estimate from memory bandwidth gives about 23 tokens/s at Q3_K_M.
- Which quantisation of ZwZ2-27B should I use on the Arc A770 16GB?
- Q3_K_M: the largest one that fits. Expected quality: noticeable loss 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.