Can the Arc Pro B60 run Gemma 4 E4B?
Yes. Gemma 4 E4B fits the Arc Pro B60's 24 GB at Q4_K_M (about 6.8 GB needed), running at an estimated 52 tok/s. There's room for up to FP16 if you want higher quality.
Every quantization
Gemma 4 E4B (8B) on 24 GB
| Quant | Weights | Needs | Fit | Est. speed |
|---|---|---|---|---|
| Q4_K_M | 5 GB | 6.8 GB | Fits easily | ~52 tok/s |
| Q5_K_M | 6 GB | 7.8 GB | Fits easily | ~44 tok/s |
| Q8_0 | 9 GB | 10.8 GB | Fits easily | ~29 tok/s |
| FP16 | 16 GB | 17.8 GB | Fits | ~16 tok/s |
Estimates. Needs = weights + 1.8 GB for KV cache and runtime at modest context. Speed is single-stream decode scaled from the Arc Pro B60's 456 GB/s bandwidth (Estimated from memory bandwidth (llama.cpp decode)). Long contexts need more memory and run slower.
Run it
Copy, paste, go
llama-server -hf google/gemma-4-E4B-it-qat-q4_0-gguf -ngl 99Ollama pulls a Q4 build by default. Pick a larger quant only if the table above says it fits. Full Gemma 4 E4B details →
Compare
Same model, other hardware - and other models on this card
Gemma 4 E4B on other hardware
- RTX 5090~206 tok/s
- RTX 5090 D~205 tok/s
- RTX 6000D~180 tok/s
- RTX 4090~125 tok/s
- RTX 3090~110 tok/s
- Radeon RX 7900 XTX~110 tok/s
- RTX 5080~110 tok/s
- RTX 3080 Ti~105 tok/s
Also runs on the Arc Pro B60
- Qwen3-Coder-30B-A3B-Instruct~99 tok/s
- Qwen3-30B-A3B-Instruct-2507~99 tok/s
- Qwen3.6 27B~16 tok/s
- Qwen3.8 27B~16 tok/s
- Gemma 4 26B A4B~77 tok/s
- gpt-oss-20b~87 tok/s
- Nemotron 3.5 Lightning~26 tok/s
- Qwen2.5-Coder-14B-Instruct~29 tok/s
Frequently asked
Quick answers
Can the Arc Pro B60 run Gemma 4 E4B?
Yes. Gemma 4 E4B fits the Arc Pro B60's 24 GB at Q4_K_M (about 6.8 GB needed), running at an estimated 52 tok/s. There's room for up to FP16 if you want higher quality.
How much VRAM does Gemma 4 E4B need?
About 5 GB of weights at Q4_K_M, 9 GB at Q8_0 and 16 GB at FP16. Add roughly 1.8 GB for the KV cache and runtime at modest context lengths.
How fast is Gemma 4 E4B on the Arc Pro B60?
Roughly 52 tokens per second at Q4_K_M (single user, llama.cpp-style decode). This is an estimate from the Arc Pro B60's 456 GB/s memory bandwidth.