Can GLiDE 27B run on the Arc A770 16GB?
Yes. At 8K context the best fit is Q3_K_M, using about 13.2 GB of the 15.2 GB usable on the Arc A770 16GB. Estimated speed ceiling: ~44 tokens/sec.
Better quality with offload: 4-bit (Q4_K_M)
Runs with offload 4-bit (Q4_K_M) at 8K context: 57 of 64 layers on the Arc A770 16GB (about 14.5 GB) and the other 7 in system RAM (about 1.7 GB). Needs a PC with at least 32 GB of system RAM. Estimated speed: 8.9–14 tokens/sec (usable for chat), versus a bandwidth ceiling of ~35 tok/s if it fit fully in VRAM.
llama-server -m glide-27b-Q4_K_M.gguf -c 8192 -ngl 57
Estimate, not a benchmark. No GGUF repo was found on Hugging Face yet; point -m at your own GGUF file. 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.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) | 10.2 GB Fits | 10.5 GB Fits | 12.0 GB Fits | 18.0 GB No | 57 |
| 3-bit (Q3_K_M) | 13.0 GB Fits | 13.2 GB Fits | 14.7 GB Tight | 20.7 GB No | 44 |
| 4-bit (Q4_K_M) | 16.0 GB No | 16.2 GB No | 17.7 GB No | 23.7 GB No | — |
| 5-bit (Q5_K_M) | 18.4 GB No | 18.7 GB No | 20.2 GB No | 26.2 GB No | — |
| 6-bit (Q6_K) | 21.0 GB No | 21.3 GB No | 22.8 GB No | 28.8 GB No | — |
| 8-bit (Q8_0) | 26.9 GB No | 27.2 GB No | 28.7 GB No | 34.7 GB No | — |
| 16-bit (F16/BF16) | 49.7 GB No | 50.0 GB No | 51.5 GB No | 57.5 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 GLiDE 27B: RX 7900 XT (20 GB, launch MSRP $899) 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
GLiDE 27B 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 (fastino/GLiDE-27B), Intel specs. Model data last checked 2026-11-18; refreshed hourly.