Build an uncensored/abliterated Greek-native LLM for local use. Best candidate: ilsp/Llama-Krikri-8B-Instruct (Llama 3.1 8B Greek fine-tune).

A user on a GMKtec EVO-X3 with an AMD Ryzen AI MAX+ 395 (Strix Halo) and 117.55 GiB unified memory is seeking guidance on building an uncensored, abliterated Greek-native LLM, naming ilsp/Llama-Krikri-8B-Instruct (a Llama 3.1 8B Greek fine-tune) as the best candidate. The plan is to abliterate the model with Heretic (heretic-llm), convert it to GGUF Q4_K_M, and run it on Vulkan via llama.cpp, with the user asking whether Heretic supports Llama-Krikri-8B, whether to abliterate safetensors before GGUF conversion or use abliterate.cpp on GGUF, and which quantization (Q4_K_M, Q5_K_M, or Q6_K) best balances Greek quality and speed. The user also asks for recommended llama.cpp Vulkan flags for Strix Halo (-ngl 99, -c 32768, flash attention, q8_0 KV cache, speculative decoding) and for any existing uncensored Greek models such as Krikri, Meltemi, or Sophea.

Hey everyone, Setup: GMKtec EVO-X3, AMD Ryzen AI MAX+ 395 Strix Halo , Radeon 8060S, 117.55 GiB unified memory, Ubuntu 24.04, Vulkan/RADV, llama.cpp built with GGML VULKAN=ON. Running Qwen3-32B-heretic-Q8 0 on llama-server 127.0.0.1:8081 . Goal: build an uncensored/abliterated Greek-native LLM for local use. Best candidate: ilsp/Llama-Krikri-8B-Instruct Llama 3.1 8B Greek fine-tune . Plan: abliterate with Heretic heretic-llm , convert to GGUF Q4 K M, run on Vulkan. Questions: 1. Has anyone abliterated Llama-Krikri-8B or any Greek fine-tune? Does Heretic support it? 2. Best workflow: abliterate safetensors then convert to GGUF, or use abliterate.cpp on GGUF? Any pitfalls? 3. Best quantization for Greek on Strix Halo: Q4 K M, Q5 K M, Q6 K? Need quality/speed balance. 4. Recommended llama.cpp Vulkan flags for Strix Halo: -ngl 99, -c 32768, flash attention, q8 0 KV cache, speculative decoding? Any known issues? 5. Any existing uncensored Greek models I missed Krikri, Meltemi, Sophea, etc. ? Thanks