Tracked topic
Qwen3 is an AI model family developed by Alibaba, released as a set of large language models for natural-language tasks.
For the first time, Qwen3.8 brings a Qwen-Max-class model to open release. Built on the architectural foundation of Qwen3.5, Qwen3.8 delivers substantial gains across coding, professional work, research, and long-horizon agentic tasks. Beyond answering harder questions, Qwen3.8 is designed to carry complex, multi-step tasks through to completion with greater reliability. Qwen3.8 features the following enhancements: Core Capabilities: Comprehensive improvements across coding, professional work, research, and long-horizon agentic tasks. Agent Execution: Stronger autonomous planning and better ha
Day 0 Support for Qwen 3.8 on AMD Instinct GPUs…AMD’s Day-0 enablement of leading open and frontier models (for example Qwen3.6, Gemma 4, Xiaomi MiMo-V2.5-Pro, and Baidu Ernie Image) means OpenMAIC can adopt new generators…
…Additional performance data for AMD RDNA4™-based GPUs in diffusion 1 and standard vision models 2 is shared below: AMD RDNA4™ GPU performance improvement over baseline Standard Vision Models: AMD RDNA4™ GPU…
…Clearly, decode performance falls off much more slowly than prefill performance. Figure 1: Llama-3.3-70B prefill and decode efficiency vs tensor parallel size Efficiency per-GPU is not the only…
…Baseline Performance with Default Compression The first evaluation was performed using Picard MarkDuplicates using the default compression setting (compression level 5) on a production whole genome public NCBI SRR359032 BAM file. For…
…To support this, AMD is expanding the AMD ROCm™ software stack with Day 0 model support, Triton performance CI, nightly Hugging Face integration, and upstream contributions to projects such as vLLM and…
…He leads the company’s high-performance server and enterprise AI business across cloud, enterprise, high-performance computing, sovereign AI and partner ecosystems. McNamara holds bachelor’s and master’s degrees in…
…Footnotes & General Disclaimer All performance and/or cost savings claims are provided by the 3rd party organization featured herein and have not been independently verified by AMD. Performance and cost benefits are…
…ZenDNN Framework landscape at a glance Performance Through ZenDNN 5.2.1, performance was measured using offline benchmarking, where a fixed batch of prompts is processed without simulating concurrent users. With ZenDNN…
…In one production deployment built around the Qwen3-235B Instruct FP8 model, developers kept the workloads running smoothly by combining the AMD ROCm software stack, vLLM, and AMD AITER, a high-performance…
To show you the most relevant results, we’ve omitted some entries very similar to those already shown. Repeat the search with the omitted results included.