The Archive

User configures dual AMD RDNA GPUs (48GB VRAM) with llama-cpp via Vulkan for local inference.

lemon-mlx-engine integrates ROCm 7.13 for AMD GPU inference of MoE and dense models on consumer hardware.

AMD announces Ryzen AI Halo developer platform and Ryzen AI Max PRO 400 series for on-device agent inference; availability details follow previous announcement.

Developer reverse-engineers PS5 APU (BC-250) register controls to unlock disabled GPU compute units for cost-effective ML inference.

AMD Ryzen AI Halo PC launches at $3,999 with 128GB onboard memory for edge AI inference.

CPU benchmark of Kokoro 82M vs Supertonic 3 TTS models on AMD EPYC; latency and quality tradeoffs measured.

User reports optimized inference performance for DeepSeek V4 Pro using ktransformers on AMD EPYC + RTX PRO hardware with tuned NUMA settings.

Shipped this for the AMD x lablab hackathon. Attached video is one of the actual reels the pipeline produced - one English sentence in, finished mp4 with characters, story, music, and voice-over out (fast demo video, not the best quality). ~45 minutes end-to-end on a single AMD Instinct MI300X. Every model is Apache 2.0 or MIT. **Pipeline (8 stages, all sequential on the same GPU):** 1. **Director Agent** - Qwen3.5-35B-A3B (vLLM + AITER MoE) plans 6 shots from one sentence, returns structured JSON with character bibles, shot prompts, music brief, per-shot voice-over script, narration langua...

Reddit discussion comparing vLLM vs llama.cpp for single-user local inference on AMD GPUs.

Luce ships DFlash+PFlash optimizations for AMD Ryzen AI MAX+ 395, achieving 2.23x decode speedup on Qwen 3.6-27B vs llama.cpp HIP.

User successfully quantized and ran DeepSeek V4 Pro locally on AMD EPYC + RTX PRO hardware using modified llama.cpp with Q4_K_M compression.

User reports 1.5–2x speedup running Qwen 27B with MTP optimization on dual AMD MI50 GPUs via llama.cpp.

Zyphra announces ZAYA1-74B-Preview, a pretraining-scale model optimized for AMD accelerators.

AMD plans PCIe-based Instinct GPUs for enterprise AI inference, expanding local LLM deployment options beyond NVIDIA.

AMD announces Instinct MI350P PCIe accelerator with CDNA 4 architecture for AI inference and training workloads.

User reports successful local inference with Qwen3.6-35B on AMD R9700 GPU, generating functional code and tests via llama-cpp.

Zyphra releases ZAYA1-8B, an 8B parameter model optimized for inference efficiency, trained on AMD hardware.

QyTw0, the Finnish AI lab founded by former AMD Silo AI CEO Peter Sarlin, is now valued at €325 million (approximately $380 million) after raising a €25 million angel round ($29 million). It's a sign of enduring tailwinds for AI, quantum computing, and sovereign tech, especially for Europe-made companies.

User reports successful MTP speculative decoding on AMD Strix Halo (AI Max 395) with llama.cpp achieving 60-80 tok/s on Qwen 3.6B GGUF.

AMD Ryzen AI Max+ 495 APU leaked with 192GB memory, enabling larger local model inference on consumer hardware.

AMD Strix Halo refresh rumored to feature 192GB+ VRAM, enabling larger MoE model inference on consumer hardware.

Community member achieves 9.7 tok/s output throughput on Kimi K2.6 with 32x AMD MI50 GPUs using vLLM fork.

AMD Halo Box reference hardware with Ryzen 395 and 128GB RAM spotted running Ubuntu; limited technical details on LLM inference capability.

AMD Ryzen 395 box launching June; likely Lenovo OEM product with no pricing details disclosed.

AMD engineers soliciting ROCm feedback from local LLM community to improve GPU acceleration stack for open-weight model inference.

Satirical post mocking AMD hardware marketing; no substantive AI news.

Hipfire local LLM dev lab acquiring full AMD GPU stack (RDNA 1-4, Strix Halo) for architecture validation and performance optimization.

HFQ4 MMQ prefill optimization for AMD Strix Halo achieves 3× speedup (310→950 tok/s) in hipfire inference engine.