Description
LocalAI version: 2.29.0
localai/localai:latest-aio-gpu-hipblast
Environment, CPU architecture, OS, and Version:
Linux xxxxx 6.14.9-300.fc42.x86_64 #1 SMP PREEMPT_DYNAMIC Thu May 29 14:27:53 UTC 2025 x86_64 GNU/Linux, Fedora 42
podman version
Client: Podman Engine
Version: 5.5.1
API Version: 5.5.1
Go Version: go1.24.3
Git Commit: 850db76dd78a0641eddb9ee19ee6f60d2c59bcfa
Built: Thu Jun 5 02:00:00 2025
Build Origin: Fedora Project
OS/Arch: linux/amd64
Describe the bug
AMD GPU detected (no GPU detected if I used the image : localai/localai:latest-gpu-hipblast )
But Is not possible to generate an image with stablediffusion the error is :
6:27PM ERR Server error error="failed to load model with internal loader: could not load model (no success): Unexpected err=RuntimeError('No HIP GPUs are available'), type(err)=<class 'RuntimeError'>
To Reproduce
podman run -ti --name local-ai -p 8085:8080 --device=/dev/kfd --device=/dev/dri/renderD128 --security-opt=no-new-privileges --group-add=video localai/localai:latest-aio-gpu-hipblas
I launched the stablediffusion for create an image same result
For me it's necessary to update rocm/hip version (In the system I used 6.3.1-4 by default)
===> LocalAI All-in-One (AIO) container starting...
AMD GPU detected
stable diffusion
Expected behavior
Used the GPGU RDNA4 (RX9070XT) acceleration
Logs
BG CPU capabilities: [3dnowprefetch abm adx aes aperfmperf apic arat avic avx avx2 bmi1 bmi2 bpext cat_l3 cdp_l3 clflush clflushopt clwb clzero cmov cmp_legacy constant_tsc cpb cpuid cqm cqm_llc cqm_mbm_local cqm_mbm_total cqm_occup_llc cr8_legacy cx16 cx8 de debug_swap decodeassists erms extapic extd_apicid f16c flushbyasid fma fpu fsgsbase fsrm fxsr fxsr_opt ht hw_pstate ibpb ibrs ibs invpcid irperf lahf_lm lbrv lm mba mca mce misalignsse mmx mmxext monitor movbe msr mtrr mwaitx nonstop_tsc nopl npt nrip_save nx ospke osvw overflow_recov pae pat pausefilter pclmulqdq pdpe1gb perfctr_core perfctr_llc perfctr_nb pfthreshold pge pku pni popcnt pse pse36 rapl rdpid rdpru rdrand rdseed rdt_a rdtscp rep_good sep sha_ni skinit smap smca smep ssbd sse sse2 sse4_1 sse4_2 sse4a ssse3 stibp succor svm svm_lock syscall tce topoext tsc tsc_scale umip user_shstk v_spec_ctrl v_vmsave_vmload vaes vgif vmcb_clean vme vmmcall vpclmulqdq wbnoinvd wdt x2apic xgetbv1 xsave xsavec xsaveerptr xsaveopt xsaves xtopology]
6:31PM DBG GPU count: 1
6:31PM DBG GPU: card #1 @0000:28:00.0 -> driver: 'amdgpu' class: 'Display controller' vendor: 'Advanced Micro Devices, Inc. [AMD/ATI]' product: 'unknown'
6:31PM DBG [startup] resolved local model: /aio/gpu-8g/embeddings.yaml
6:31PM DBG [startup] resolved local model: /aio/gpu-8g/rerank.yaml
6:31PM DBG [startup] resolved local model: /aio/gpu-8g/text-to-speech.yaml
6:31PM DBG [startup] resolved local model: /aio/gpu-8g/image-gen.yaml
6:31PM DBG [startup] resolved local model: /aio/gpu-8g/text-to-text.yaml
6:31PM DBG [startup] resolved local model: /aio/gpu-8g/speech-to-text.yaml
6:31PM DBG [startup] resolved local model: /aio/gpu-8g/vad.yaml
6:31PM DBG [startup] resolved local model: /aio/gpu-8g/vision.yaml
6:33 PM ERR Server error error="failed to load model with internal loader: could not load model (no success): Unexpected err=RuntimeError('No HIP GPUs are available'), type(err)=<class 'RuntimeError'>
Additional context
I test with another Radeon (RX6500XT) at next moment for confirm if it's ok or not.