Robert Važan

TryFrom is however still advertised in [the latest docs](https://docs.rs/strum/latest/strum/derive.FromRepr.html), which made me wonder why it does not work. BTW, I could use TryFrom myself, perhaps as an optional feature. Or...

@nagavenkateshgavini The problem still exists, although I have found workarounds for my use cases.

Latest (0.1.27) docker image with ROCm works for me on Ryzen 5600G with 8GB VRAM allocation. Prompt processing is 2x faster than with CPU. Generation runs at max speed even...

See also discussion in the #738 epic.

> ROCm unsupported integrated GPU detected Ollama skipped the iGPU, because it has less than 1GB of VRAM. You have to configure VRAM allocation for the iGPU in BIOS to...

> But normal behaviour for the iGPU should be that it requests more VRAM if needed. Why do you think so? Where is it documented? Mine maxes at 512MB unless...

iGPUs indeed do allocate system RAM on demand. It's called [GTT/GART](https://en.wikipedia.org/wiki/Graphics_address_remapping_table). Here's what I get when I run `sudo dmesg | grep "M of"` on my system with 32GB RAM:...

@chiragkrishna Do you mean this? https://github.com/ggerganov/llama.cpp/pull/4449 Since llama.cpp already supports UMA (GGT/GART), Ollama could perhaps include llama.cpp build with UMA enabled and use it when the conditions are right (AMD...

@DocMAX I don't think there's UMA support in ollama yet. It's a compile-time option in llama.cpp. The other env variables (HSA_OVERRIDE_GFX_VERSION was sufficient in my experiments) are correctly passed down...

Not yet. It's in the TODO list though.