https://gpufit.tech/gpu/arc-a380/New catalog models scored against Intel Arc A380.https://gpufit.tech/models/nvidia-nemotron-3-5-bolt-e4b/https://gpufit.tech/models/nvidia-nemotron-3-5-bolt-e4b/Fri, 23 Oct 2026 09:47:55 GMTRuns fully on your Intel Arc A380 at UD-Q3_K_XL, 8K context, about 21–28 tok/s. Weights 3.93 GB plus 0.14 GB of KV cache — 4.41 GB of 5.22 GB usable. Up to 32K context still fits. 21–28 tok/s.https://gpufit.tech/models/deepseek-v4-1-flash-1017/https://gpufit.tech/models/deepseek-v4-1-flash-1017/Sat, 17 Oct 2026 20:36:06 GMTWon't fit on Intel Arc A380. The smallest quant UD-IQ1_S is 217.0 GB before KV cache, and usable memory is 5.22 GB.https://gpufit.tech/models/lfm2-5-24b-a2b/https://gpufit.tech/models/lfm2-5-24b-a2b/Tue, 13 Oct 2026 14:19:48 GMTExpert offload on your Intel Arc A380: UD-Q4_K_XL keeps shared weights on the GPU and about 11.5 GB of experts in system RAM. About 9.7–13 tok/s. Active-parameter speed is approximate. 9.7–13 tok/s.