01 // Representative inference estimates
Single-stream decode · llama.cpp
Aggregated from published community sources. Test setups differ, so use close results as directional evidence.
Sources and normalization →01b // Performance across quantization
vs. nearest competitors
How tok/s scales from FP16 → Q8 → Q4 compared to GPUs in a similar price/VRAM range.
02 // Hardware specs
ArchitectureXe2-HPG
Process nodeTSMC 4nm
Memory12 GB
Memory bandwidth456 GB/s
FP16 compute14.6 TFLOPS
INT8 compute29 TOPS
TDP190 W
PCIeGen 4 x8
Form factorDual-slot
CoolingAxial
03 // Model fit
Weight estimate plus 15% planning headroom for runtime buffers and a modest KV cache. Long context can require more.
+ STRENGTHS
- ✓12GB clears our Qwen3 14B Q4 planning target
- ✓456 GB/s memory bandwidth · top tier in its class
- ✓Indexed formats: FP16, Q8, Q4 · verify support in your runtime
− TRADE-OFFS
- −Draws 190W under load — plan PSU and thermals accordingly
- −Limited to dual-slot chassis
- −Driver lock-in to vendor stack
related research
Research behind Intel Arc B580 inference tradeoffs
These papers explain the quantization, cache, bandwidth, and runtime constraints that matter before buying this GPU for local AI.
GPU inference optimization papers
Memory bandwidth, FlashAttention, dequant kernels, and backend maturity.
Open Local AI inference papers
llama.cpp, Apple Silicon, constrained GPUs, offload, and one-box inference.
Open LLM serving systems papers
vLLM, PagedAttention, speculative decoding, batching, and GPU servers.
Open