2 × Ryzen AI Max+ 395 LLM benchmarks
10 results · 10 measurements · 2 sources
Device specifications
| Variant | Memory | Memory type | Bus | Bandwidth | FP16 matrix | INT8 matrix |
|---|---|---|---|---|---|---|
| Ryzen AI Max+ 395 (Radeon 8060S) ↗ | SharedUp to 128 GB system RAM | LPDDR5x | 256-bit | — | — | — |
Manufacturer specifications. Shared system RAM is not dedicated VRAM. Unreported values remain —; measured LLM results retain their reported configuration.
Selected reported tests
10 of 10 resultsSource-reported results, selected across reports by documentation quality. Match configurations before comparing values.
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | 484.53 tok/sPrefill · mean · per request | — | 2 × Ryzen AI Max+ 395Multi-GPU · split unknown · Unified memory · 256 GB RAMConfiguration
|
|---|---|---|---|---|---|---|
| View ReportGeneric LLM | Gemma4-26B-CIRU-AMD-STRIX-vLLM | Token speed | 661.74 tok/sThroughput · aggregate | — | — | 2 × Ryzen AI Max+ 395Independent replicas · Unified memory · 128 GB RAMConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | — | 25.79 tok/sGeneration · mean · per request | 2 × Ryzen AI Max+ 395Multi-GPU · split unknown · Unified memory · 256 GB RAMConfiguration
|
| View ReportGeneric LLM | Gemma4-26B-CIRU-AMD-STRIX-vLLM | Token speed | 1,165.98 tok/sThroughput · peak · aggregate | — | — | 2 × Ryzen AI Max+ 395Independent replicas · Unified memory · 128 GB RAMConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | 487.97 tok/sPrefill · mean · per request | — | 2 × Ryzen AI Max+ 395Multi-GPU · split unknown · Unified memory · 256 GB RAMConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | — | 26.03 tok/sGeneration · mean · per request | 2 × Ryzen AI Max+ 395Multi-GPU · split unknown · Unified memory · 256 GB RAMConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | 486.77 tok/sPrefill · mean · per request | — | 2 × Ryzen AI Max+ 395Multi-GPU · split unknown · Unified memory · 256 GB RAMConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | — | 26.31 tok/sGeneration · mean · per request | 2 × Ryzen AI Max+ 395Multi-GPU · split unknown · Unified memory · 256 GB RAMConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | 489.61 tok/sPrefill · mean · per request | — | 2 × Ryzen AI Max+ 395Multi-GPU · split unknown · Unified memory · 256 GB RAMConfiguration
|
| View Reportllama-bench | Qwen3-Coder-Next | Token speed | — | — | 26.27 tok/sGeneration · mean · per request | 2 × Ryzen AI Max+ 395Multi-GPU · split unknown · Unified memory · 256 GB RAMConfiguration
|
Models and software in this collection
Model families: Gemma 4, Qwen3
Engines: llama.cpp, vLLM (Ciru modified)
Methodology and evidence standards · Download the research dataset