Can Surface Laptop Ultra · RTX Spark N1X (20-core) · 128 GB run Reflection Beam (501B-A23B)?
128 GB unified · 300 GB/s memory bandwidth · Reflection Beam (501B-A23B) has 501B parameters (23B active per token).
Short answer
No. This combination isn't supported.
Reflection Beam (501B-A23B) is not published in a format llama.cpp (llama-server) can load (GGUF).
The same applies to every workload we rate (chat, coding questions, coding in a repository, agentic coding and long documents): they all need at least as much memory as chat.
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 Reflection Beam (501B-A23B)
Hardware rated comfortable or better for chat, lowest price first.
- Mac Studio M5 Ultra 512 GB$7,099 · 25–40 tok/sExcellent
- Mac Studio M3 Ultra 512 GB$7,099 · 17–25 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.
Open sourceCheck my exact setup