Can Surface Laptop Ultra · RTX Spark N1X (20-core) · 128 GB run Llama 4 Scout (109B-A17B)?
128 GB unified · 300 GB/s memory bandwidth · Llama 4 Scout (109B-A17B) has 108.6B parameters (17B active per token).
Short answer
Yes, comfortable for chat; borderline for agentic coding.
Best quantization for chat: Q4_K_M (61 GB of weights) on Ollama. Generation: 10–17 tok/s, leaving about 53 GB free for your other apps.
How it rates for each workload
Same model and machine, different jobs. Agents and long documents send far bigger prompts than chat, so they need much more speed and memory.
| Workload | Verdict | Quant | Generation | First reply | Context | |
|---|---|---|---|---|---|---|
| Chatwith Open WebUI | Comfortable | Q4_K_M | 10–17 tok/s | under 1 s | 8K | Details |
| Coding questionswith Continue | Acceptable | Q5_K_M | 8–14 tok/s | ≈4 s | 16K | Details |
| Coding in a repositorywith Aider | Acceptable | Q5_K_M | 7–12 tok/s | ≈10 s | 32K | Details |
| Agentic codingwith OpenCode | Borderline | Q6_K | 6–10 tok/s | ≈15 s | 32K | Details |
| Long documents (64K)with Open WebUI | Acceptable | Q4_K_M | 5–9 tok/s | ≈60 s | 64K | Details |
Speeds are bandwidth-based estimates; no direct benchmark exists for this pair. “First reply” is the time to read the workload’s opening prompt. How we calculate
Get it running
Copy-paste commands for the Q4_K_M build with a 64K context. Ollama and llama.cpp download the file straight from Hugging Face.
ollama run hf.co/unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF:Q4_K_M
This build is split into 2 files, which Ollama can't import from Hugging Face. Use llama.cpp or LM Studio instead.
Other models for Surface Laptop Ultra · RTX Spark N1X (20-core) · 128 GB
The best-rated models for chat on this machine.
- gpt-oss-120b40–65 tok/sExcellent
- Qwen3.6 35B-A3B45–75 tok/sExcellent
- Qwen3-Next 80B-A3B Instruct45–75 tok/sExcellent
- Gemma 4 26B-A4B35–60 tok/sExcellent
- LFM2.5 24B-A2B85–150 tok/sExcellent
- Qwen3.5 35B-A3B45–75 tok/sExcellent
Cheapest machines for Llama 4 Scout (109B-A17B)
Hardware rated comfortable or better for chat, lowest price first.
- Ryzen AI Max+ 395 mini PC / Framework Desktop · 128 GB$2,000 · 11–16 tok/sComfortable
- Mac Studio M2 Max 96 GB$3,799 · 18–25 tok/sExcellent
- MacBook Pro M3 Max 96 GB$3,999 · 14–19 tok/sComfortable
- Mac Studio M5 Max 128 GB$4,099 · 20–35 tok/sExcellent
- Mac Studio M4 Max 128 GB$4,099 · 25–30 tok/sExcellent
- MacBook Pro M2 Max 96 GB$5,099 · 18–25 tok/sExcellent
Same chip, other memory: 24 GB
Different tool, context size or workload?
These pages use typical settings. The full check models your exact setup.