Can Muse Glimmer 30B run on the Arc A770 16GB?
Yes. At 8K context the best fit is Q2_K, using about 11.4 GB of the 15.2 GB usable on the Arc A770 16GB. Estimated speed ceiling: ~50 tokens/sec.
Better quality with offload: 4-bit (Q4_K_M)
Runs with offload 4-bit (Q4_K_M) at 8K context: 42 of 52 layers on the Arc A770 16GB (about 14.6 GB) and the other 10 in system RAM (about 3.3 GB). Needs a PC with at least 32 GB of system RAM. Estimated speed: 5.3–9.1 tokens/sec (slow but usable), versus a bandwidth ceiling of ~31 tok/s if it fit fully in VRAM.
llama-server -hf bartowski/Muse-Glimmer-30B-GGUF:Q4_K_M -c 8192 -ngl 42 --no-mmproj
Estimate, not a benchmark. The command downloads the GGUF from Hugging Face. The speed range spans dual-channel DDR4-3200 to DDR5-6000 system RAM; your CPU, RAM speed and llama.cpp version move it. If llama.cpp runs out of GPU memory, lower -ngl by 2-3; if VRAM is left over, raise it for more speed. Recent llama.cpp builds can also pick the split automatically. How we estimate offload.
+ 3.6 GB with image input. The figures on this page are for text only. With image input enabled, the vision projector (mmproj) file loads alongside the model: Q2_K at 8K context then needs about 14.9 GB (tight). llama.cpp's -hf downloads and loads it automatically unless you add --no-mmproj. Calculator with image input on.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) | 11.3 GB Fits | 11.4 GB Fits | 11.7 GB Fits | 12.9 GB Fits | 50 |
| 3-bit (Q3_K_M) | 14.5 GB Tight | 14.5 GB Tight | 14.8 GB Tight | 16.0 GB No | 38 |
| 4-bit (Q4_K_M) | 17.9 GB No | 17.9 GB No | 18.2 GB No | 19.4 GB No | — |
| 5-bit (Q5_K_M) | 20.7 GB No | 20.7 GB No | 21.0 GB No | 22.2 GB No | — |
| 6-bit (Q6_K) | 23.7 GB No | 23.7 GB No | 24.0 GB No | 25.2 GB No | — |
| 8-bit (Q8_0) | 30.4 GB No | 30.4 GB No | 30.7 GB No | 32.0 GB No | — |
| 16-bit (F16/BF16) | 56.4 GB No | 56.4 GB No | 56.7 GB No | 57.9 GB No | — |
• real GGUF file size. Speed is a bandwidth ceiling (560 GB/s), not a benchmark. Methodology.
What fits instead
- Cheapest listed GPU that runs Muse Glimmer 30B: RX 7900 XTX (24 GB, launch MSRP $999) at Q4_K_M, fully in VRAM at 8K context.
- Largest popular model that fits the Arc A770 16GB: gpt-oss 20B (21B, MoE) at Q8_0, about 12.3 GB at 8K context.
"Runs" and "fits" mean the whole model in GPU memory at 8K context (4-bit or better for the model pick). Prices are US launch MSRPs, not today's street prices. "Popular" means the 60 most-downloaded models we track.
More
Muse Glimmer 30B on every GPU and Mac · Every model that runs on the Arc A770 16GB · Other models that fit 16 GB VRAM · Try other settings in the calculator
Guides: Best GPU for local LLMs by budget · Can I run gpt-oss, Llama 70B or Qwen3? · Mac vs PC for local LLMs
Data: Hugging Face Hub (meta-models/Muse-Glimmer-30B), Intel specs. Model data last checked 2026-11-18; refreshed hourly.