AMD Ryzen AI Max+ 395 vs Intel Core Ultra 9 285K

Agent access

1. ai-native userPoint an agent at llms.txt or agent-oriented docs

   weight 2 · round to AMD Ryzen AI Max+ 395
   AMD Ryzen AI Max+ 395✓fullprobed7/10

   A direct probe confirms that AMD's Ryzen AI developer docs serve a working llms.txt file (HTTP 200 with structured content) at ryzenai.docs.amd.com, which an AI agent could be pointed at directly. Missing for 10: broader agent-oriented documentation structure beyond the single llms.txt file, and independent/community confirmation of agents actually using it.

   • [probe] “PROBE llms.txt: HTTP 200 at https://ryzenai.docs.amd.com/llms.txt # Ryzen AI Software > Note: ROCm documentation is split across multiple p…”
   • [probe] “PROBE runtime (recorded 2026-09-15): the Ryzen AI Software documentation at ryzenai.docs.amd.com answered a keyless curl naming Ryzen AI — t…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K~partialprobed5/10

   Intel's corporate site does serve a machine-fetchable llms.txt (confirmed via probe returning HTTP 200 with a short company description), so an agent could be pointed at it, but the snippet is generic corporate boilerplate with no product-specific or agent-oriented documentation for the Core Ultra 9 285K itself, and no openapi/agent API endpoints were found. missing for 10: richer llms.txt content tailored to product docs, presence of llms-full.txt or structured agent-facing spec docs, independent confirmation the file is kept current.

   • [probe] “PROBE llms.txt: HTTP 200 at https://www.intel.com/llms.txt # Intel Corporation > Intel is a global technology company delivering AI compute…”
   • [probe] “PROBE openapi: all candidate paths 404 (https://www.intel.com/openapi.json, https://www.intel.com/swagger.json, https://www.intel.com/api/op…”

   proof ↗⚑ flag

2. ai-native userRun the product headlessly / in CI for automation

   weight 2 · round to AMD Ryzen AI Max+ 395
   AMD Ryzen AI Max+ 395~partialclaimed4/10

   The Ryzen AI software stack supports headless Linux server OSes (Ubuntu, RHEL) and exposes CLI/API-driven inference via ONNX Runtime (C++/Python APIs) and the Lemonade SDK for llama.cpp/OGA, which are automatable without a GUI. However, there is no explicit documentation of CI pipelines, containerized/Docker deployment, or automated build/test workflows for this hardware. Missing for 10: explicit CI/automation examples, containerization support, and independent confirmation of headless scripted runs.

   • [claimed-docs] “The AI model is deployed using the ONNX Runtime with either C++ or Python APIs.”
   • [claimed-docs] “the Lemonade SDK, which is multi-vendor open-source software that provides everything necessary for quickly getting started with LLMs on OGA…”
   • [claimed-docs] “Windows 11 - 64-Bit Edition , RHEL x86 64-Bit , Ubuntu x86 64-Bit”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K—none0/10

   The axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na". (na/none harmonized at arena bring-up — see pipeline/scripts/na-harmonize.ts.)

   ⚑ flag

3. ai-native userDrive the product through a documented public API

   weight 3 · round to AMD Ryzen AI Max+ 395
   AMD Ryzen AI Max+ 395~partialprobed4/10

   AMD documents software-level APIs (ONNX Runtime C++/Python APIs, Vitis AI EP, Lemonade SDK) for driving AI workloads on the NPU, but there is no public REST/OpenAPI-style programmatic interface for 'driving the product' as an agentic system — probes for OpenAPI/swagger specs all 404. missing for 10: a documented public REST/agent-facing API or SDK entry point beyond ML framework runtimes, independent confirmation of API usage for agentic control.

   • [claimed-docs] “The AI model is deployed using the ONNX Runtime with either C++ or Python APIs.”
   • [claimed-docs] “the Lemonade SDK, which is multi-vendor open-source software that provides everything necessary for quickly getting started with LLMs on OGA…”
   • [probe] “PROBE openapi: all candidate paths 404 (https://ryzenai.docs.amd.com/openapi.json, https://ryzenai.docs.amd.com/swagger.json, https://ryzena…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K—none0/10

   The axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na". (na/none harmonized at arena bring-up — see pipeline/scripts/na-harmonize.ts.)

   ⚑ flag

4. ai-native userBuild against official SDKs

   weight 2 · round to AMD Ryzen AI Max+ 395
   AMD Ryzen AI Max+ 395✓fullprobed7/10

   AMD provides official Ryzen AI SDK docs covering ONNX Runtime with C++/Python APIs, the Vitis AI EP, AMD Quark quantization toolkit, and the Lemonade SDK for LLMs, all live and crawlable at ryzenai.docs.amd.com, giving AI-native developers concrete official SDKs to build against. Missing for 10: independent hands-on developer reports validating SDK usability/completeness, and no OpenAPI/formal API reference confirmed (probe found 404s for openapi endpoints).

   • [claimed-docs] “This allows developers to build and deploy models trained in PyTorch or TensorFlow and run them directly on laptops powered by Ryzen AI usin…”
   • [claimed-docs] “AMD Quark is a comprehensive cross-platform deep learning toolkit designed to simplify and enhance the quantization of deep learning models.”
   • [claimed-docs] “The AI model is deployed using the ONNX Runtime with either C++ or Python APIs.”
   • [claimed-docs] “the Lemonade SDK, which is multi-vendor open-source software that provides everything necessary for quickly getting started with LLMs on OGA…”
   • [probe] “PROBE llms.txt: HTTP 200 at https://ryzenai.docs.amd.com/llms.txt # Ryzen AI Software > Note: ROCm documentation is split across multiple p…”
   • [probe] “PROBE runtime (recorded 2026-09-15): the Ryzen AI Software documentation at ryzenai.docs.amd.com answered a keyless curl naming Ryzen AI — t…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K~partialclaimed6/10

   Intel documents official SDKs and toolkits (oneAPI, oneDNN, DPC++ Compatibility Tool, VTune Profiler) and lists supported AI frameworks (OpenVINO, DirectML, ONNX RT, WebNN) for the Core Ultra platform, which an AI-native developer could build against. However, this evidence is generic to the Core Ultra line rather than specific hands-on validation for the 285K, and there's no independent corroboration of developer experience with these SDKs. missing for 10: independent/hands-on validation of SDK usability, concrete code examples or developer testimonials, and confirmation these tools are specifically exercised on the 285K rather than just documented for the broader Core Ultra family.

   • [claimed-docs] “Maximize AI PC inference capabilities from large language models (LLM) to image generation with Intel® oneAPI Deep Neural Network Library (o…”
   • [claimed-docs] “Optimize performance on client GPUs and NPUs from Intel with new analysis tool features in Intel® VTune™ Profiler.”
   • [claimed-docs] “Intel® DPC++ Compatibility Tool for CUDA-to-SYCL migration”
   • [claimed-docs] “AI Software Frameworks Supported by CPU OpenVINO™, WindowsML, DirectML, ONNX RT, WebNN”

   proof ↗⚑ flag

Agentic features

1. ai-native userOperate the product with natural-language commands

   weight 2 · round to Intel Core Ultra 9 285K
   AMD Ryzen AI Max+ 395—none0/10

   The axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na". (na/none harmonized at arena bring-up — see pipeline/scripts/na-harmonize.ts.)

   ⚑ flag

   Intel Core Ultra 9 285K~partialclaimed3/10

   Marketing copy claims 'voice commands to get answers, edit content, and turn ideas into reality,' suggesting some natural-language interaction on Core Ultra platforms, but this is a single vague marketing line with no product specifics, no named assistant, and no hands-on corroboration. Missing for 10: concrete details on which software/assistant enables this, independent verification it works as described, and evidence of broader NL command support beyond voice.

   • [claimed-docs] “Use voice commands to get answers, edit content, and turn ideas into reality.”

   proof ↗⚑ flag

Heavy compute

1. developerCompile big codebases and run heavy parallel jobs fast — documented core counts and boost behavior, corroborated by independent multi-core benchmarks

   weight 3 · round to Intel Core Ultra 9 285K
   AMD Ryzen AI Max+ 395—none0/10

   No evidence pack item documents core counts, boost clock behavior, or independent multi-core/compile benchmarks; the AMD docs and product page instead focus on AI/NPU software stack, TDP, and I/O specs. Community comments (comm-2) even push back on judging this chip by CPU-only benchmarks, but no concrete multi-core compile benchmark or core/boost spec is cited either way.

   ⚑ flag

   Intel Core Ultra 9 285K~partialprobed6/10

   Intel's ARK spec page (confirmed live via probe) documents core counts and boost clocks, and independent reviews (Tom's Hardware, Phoronix, HN discussion) corroborate strong multi-threaded productivity performance versus its predecessor, supporting compile/parallel-job workloads. However, the same independent sources show AMD's 9950X ahead in raw multi-core performance/perf-per-watt, and Phoronix reports real compiler segfaults tied to non-rated RAM speeds — a concrete caveat for compiling workloads. missing for 10: explicit quoted core-count/boost-clock figures in the evidence pack itself, quantitative independent multi-core benchmark numbers (only qualitative 'productivity performance' claims), and confirmation the RAM-related compiler stability issue is fully resolved across configurations.

   • [probe] “PROBE runtime (recorded 2026-09-15): Intel's ARK specifications page for the Core Ultra 9 285K answered a keyless curl and names the part — …”
   • [community] “Intel's Core Ultra 9 285K makes strong gains in productivity workloads, but it struggles to match its prior-gen counterpart in gaming perfor…”
   • [community] “Pros: Productivity performance, Power consumption and efficiency, Support for CUDIMM memory, Relaxed cooling requirements, Higher memory OC …”
   • [community] “Compared to its predecessor, 14900K, it ends up both faster (barring some outliers) and significantly more power efficient. Compared to Zen …”
   • [community] “When running with Corsair DDR5-8000 DIMMs, I was encountering compiler segmentation faults occasionally... With some DDR5-6000 DIMMs I also …”
   • [community] “So, very, very similar to the 9950X score (100.x%)... But without AVX512 and possibly much higher power consumption.”

   proof ↗⚑ flag

Single thread

1. power-userEveryday interactive work feels instant — leading single-thread performance shown in independent benchmarks, not just a peak-GHz number on a slide

   weight 2 · round to Intel Core Ultra 9 285K
   AMD Ryzen AI Max+ 395—none0/10

   No independent single-thread CPU benchmark data appears anywhere in the evidence pack; the only performance-related community comments concern GPU/LLM token throughput and memory bandwidth, and one explicitly notes the CPU is not the reason to choose this part ('If you're just going to use the CPU obviously the 395 is not what you want'). There's no vendor or third-party single-thread benchmark citation to support the story.

   • [community] “Terrible benchmarks. They should probably compare the 395 GPU against the 9950 CPU or against a 7600/9060. If you're just going to use the C…”
   • [community] “Own a framework desktop 128gb. GPU bandwidth limits the amount of tokens/sec that you get. I've mostly been running QWEN-3.6-35B-A3B at Q8 a…”
   • [community] “With 128 GB of unified memory, large models can fit into memory, but memory bandwidth quickly becomes the limiting factor. With long context…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K!disputedcontradicted4/10

   Intel markets the 285K on flashy specs and AI/creative workloads, but independent reviews show single-thread/gaming performance actually regressed vs the prior-gen 14900K and lags AMD's 9950X/7800X3D, with Windows-specific single-threaded performance issues reported at launch. missing for 10: independent benchmark data showing leading single-thread performance, resolution of the documented single-thread/gaming regressions, third-party corroboration of 'instant' interactive responsiveness.

   • [claimed-docs] “Find your flow state. It’s even easier to create, edit and render with the programs you love, on a platform you already know.”
   • [community] “Intel's Core Ultra 9 285K makes strong gains in productivity workloads, but it struggles to match its prior-gen counterpart in gaming perfor…”
   • [community] “Compared to its predecessor, 14900K, it ends up both faster (barring some outliers) and significantly more power efficient. Compared to Zen …”
   • [community] “In Windows reviews, 285K's performance is even worse, particularly in gaming tests where it is slower than 14900K and 7800X3D... 'When pairi…”
   • [community] “Arrow lake has had quite a messy launch on Windows. Intel, Microsoft, and motherboard makers have made a few changes to prevent crashes and …”

   proof ↗⚑ flag

Toolchain

1. developerThis architecture is a first-class development target — mature compilers, official optimization guidance, and an OS and tooling ecosystem that treats it as tier one

   weight 3 · round drawn
   AMD Ryzen AI Max+ 395~partialcommunity5/10

   AMD provides official Ryzen AI docs, ONNX Runtime + Vitis AI EP, the Quark quantization toolkit, and the Lemonade SDK, plus OS support for Windows 11, RHEL, and Ubuntu, indicating a real first-party tooling stack. However, community reports show developers relying on third-party runtimes (llama.cpp/Vulkan, Ollama) rather than a mature native compiler/optimization path, and cite memory-bandwidth bottlenecks and confusing benchmarks/naming as friction points, suggesting the ecosystem is still maturing rather than unambiguously tier-one. Missing for 10: evidence of mature first-party compiler toolchains, independent benchmarks confirming optimization guidance efficacy, and confirmation that NPU/GPU stack is treated as tier-one by major ML frameworks beyond ONNX.

   • [claimed-docs] “This allows developers to build and deploy models trained in PyTorch or TensorFlow and run them directly on laptops powered by Ryzen AI usin…”
   • [claimed-docs] “AMD Quark is a comprehensive cross-platform deep learning toolkit designed to simplify and enhance the quantization of deep learning models.”
   • [claimed-docs] “the Lemonade SDK, which is multi-vendor open-source software that provides everything necessary for quickly getting started with LLMs on OGA…”
   • [claimed-docs] “Windows 11 - 64-Bit Edition , RHEL x86 64-Bit , Ubuntu x86 64-Bit”
   • [community] “Own a framework desktop 128gb. GPU bandwidth limits the amount of tokens/sec that you get. I've mostly been running QWEN-3.6-35B-A3B at Q8 a…”
   • [community] “I'm running qwen3.6, gpt-oss and embeddinggemma with Ollama on Ryzen 7 8700G + 96 GB DDR5 with 12-14 tokens/sec... On Strix Halo it will run…”
   • [community] “With 128 GB of unified memory, large models can fit into memory, but memory bandwidth quickly becomes the limiting factor. With long context…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K~partialcommunity5/10

   Intel provides substantive official developer tooling (oneAPI, VTune Profiler, DPC++ CUDA-to-SYCL migration, oneDNN/PyTorch optimizations, OpenVINO/DirectML/ONNX support) suggesting tier-one treatment for AI/HPC workloads. However, community evidence shows real platform-maturity friction: compiler segmentation faults tied to memory configuration, a 'messy' Arrow Lake Windows launch requiring OS/firmware fixes, and reports of severe performance regressions on newer Windows builds — undermining the 'mature, tier-one, no caveats' framing. Missing for 10: independent benchmarks confirming stable toolchain behavior across compilers (GCC/MSVC/LLVM) without workarounds, evidence of Linux distro-level tier-one support, and resolution confirmation for the reported crashes/perf regressions.

   • [claimed-docs] “Maximize AI PC inference capabilities from large language models (LLM) to image generation with Intel® oneAPI Deep Neural Network Library (o…”
   • [claimed-docs] “Optimize performance on client GPUs and NPUs from Intel with new analysis tool features in Intel® VTune™ Profiler.”
   • [claimed-docs] “Intel® DPC++ Compatibility Tool for CUDA-to-SYCL migration”
   • [claimed-docs] “AI Software Frameworks Supported by CPU OpenVINO™, WindowsML, DirectML, ONNX RT, WebNN”
   • [community] “When running with Corsair DDR5-8000 DIMMs, I was encountering compiler segmentation faults occasionally... With some DDR5-6000 DIMMs I also …”
   • [community] “Arrow lake has had quite a messy launch on Windows. Intel, Microsoft, and motherboard makers have made a few changes to prevent crashes and …”
   • [community] “In Windows reviews, 285K's performance is even worse, particularly in gaming tests where it is slower than 14900K and 7800X3D... 'When pairi…”

   proof ↗⚑ flag

Virtualization

1. developerVMs and containers run well on this silicon — documented virtualization support and mainstream hypervisor and Docker workflows

   weight 2 · round drawn
   AMD Ryzen AI Max+ 395—none0/10

   The evidence pack lists supported OSes (Windows 11, RHEL, Ubuntu) but contains no documentation of virtualization features (SVM/AMD-V, IOMMU), hypervisor compatibility (KVM/Hyper-V/VMware), or Docker/container workflows on this chip. Virtualization support is a fair question for a CPU/SoC but no evidence confirms it here.

   • [claimed-docs] “Windows 11 - 64-Bit Edition , RHEL x86 64-Bit , Ubuntu x86 64-Bit”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K—none0/10

   The evidence pack contains no documentation of VT-x/VT-d virtualization extensions, hypervisor compatibility (Hyper-V, KVM, VMware), or container/Docker workflows for the 285K — only general AI/gaming/creator marketing, vPro and ECC specs, and community performance reviews unrelated to virtualization.

   ⚑ flag

Gaming fps

1. gamerThis chip drives high frame rates in real games — vendor gaming claims (cache, boost behavior, integrated GPU class) corroborated by independent game benchmarks

   weight 3 · round to Intel Core Ultra 9 285K
   AMD Ryzen AI Max+ 395—none0/10

   The evidence pack contains no vendor gaming claims (cache/boost/iGPU class positioning) and no independent game benchmark data or FPS figures; community comments are about AI/LLM token throughput and general benchmark methodology complaints, not real-game performance.

   • [community] “Terrible benchmarks. They should probably compare the 395 GPU against the 9950 CPU or against a 7600/9060. If you're just going to use the C…”
   • [community] “It would have been interesting to see the graphics performance and AI performance of the 395 compared to discrete gpus... how does it stack …”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K!disputedcontradicted3/10

   Intel's marketing touts smooth FPS and GPU-class gaming features, but independent reviews directly contradict this for the 285K desktop chip: Tom's Hardware reports 'generational regression in gaming performance' making it worse than the prior-gen 14900K and worse value than AMD, and HN citations confirm it trails both the 14900K and 7800X3D in gaming benchmarks, with some Windows configs seeing 50% FPS drops. Missing for 10: any vendor-cited gaming benchmark specific to the 285K, and any independent benchmark showing it competitively winning frame-rate comparisons.

   • [community] “Intel's Core Ultra 9 285K makes strong gains in productivity workloads, but it struggles to match its prior-gen counterpart in gaming perfor…”
   • [community] “Pros: Productivity performance, Power consumption and efficiency, Support for CUDIMM memory, Relaxed cooling requirements, Higher memory OC …”
   • [community] “In Windows reviews, 285K's performance is even worse, particularly in gaming tests where it is slower than 14900K and 7800X3D... 'When pairi…”
   • [claimed-docs] “More GPU performance from Intel® Arc™ graphics and AI-enhanced gameplay with Intel® XeSS 3 give you ultra-smooth FPS on the go.”

   proof ↗⚑ flag

Media engines

1. creatorHardware media engines carry my editing and streaming — documented hardware encode and decode (AV1, HEVC, ProRes-class) on the chip itself

   weight 2 · round to Intel Core Ultra 9 285K
   AMD Ryzen AI Max+ 395—none0/10

   Evidence pack contains no documentation of hardware media/video encode-decode engines (AV1, HEVC, ProRes-class) on this chip; all citations focus on AI/NPU software stack, CPU/GPU specs, TDP, and community discussion of LLM performance. Missing for 10: any mention of media/video codec engine, encode/decode capability, or streaming/editing hardware acceleration documentation.

   ⚑ flag

   Intel Core Ultra 9 285K~partialclaimed3/10

   Intel's marketing mentions an integrated 'Xe Media Engine' enabling simultaneous play/stream/edit with 'sharp, high-def video' and general Arc graphics support for editing/rendering, but the evidence pack never names specific codec support (AV1, HEVC, ProRes-class) for encode/decode on this chip. missing for 10: explicit documentation of AV1/HEVC hardware encode+decode, ProRes support, and any independent benchmark/hands-on verification of media engine performance.

   • [claimed-docs] “Play, stream, and edit, all at once. Intel Xe Media Engine delivers sharp, high-def video without compromising game graphics.”
   • [claimed-docs] “Tackle 3D models, long edits, and heavy exports with Intel® Arc™ graphics, even unplugged and on-the-go.”
   • [claimed-docs] “With 50% more graphics cores on select SKUs, you get the speed to game, stream and edit media with crisp visuals”

   proof ↗⚑ flag

Npu sdk

1. ai-native userThe chip's AI acceleration is exposed to developers — an NPU or neural engine with a published TOPS figure (precision stated) and an official SDK or runtime that ships today

   weight 3 · round to AMD Ryzen AI Max+ 395
   AMD Ryzen AI Max+ 395~partialprobed6/10

   Strong evidence of an official, shipping SDK/runtime stack (Ryzen AI Software docs, ONNX Runtime + Vitis AI Execution Provider, AMD Quark quantization toolkit, Lemonade SDK for LLMs) confirming real developer-facing NPU acceleration tooling. However, the evidence pack contains no published TOPS figure or precision spec for the NPU itself, which the story explicitly requires. Missing for 10: a documented TOPS number with precision (e.g., INT8/INT4) for the XDNA2 NPU, and independent benchmark corroboration of that figure.

   • [claimed-docs] “This allows developers to build and deploy models trained in PyTorch or TensorFlow and run them directly on laptops powered by Ryzen AI usin…”
   • [claimed-docs] “AMD Quark is a comprehensive cross-platform deep learning toolkit designed to simplify and enhance the quantization of deep learning models.”
   • [claimed-docs] “The AI model is deployed using the ONNX Runtime with either C++ or Python APIs.”
   • [claimed-docs] “the Lemonade SDK, which is multi-vendor open-source software that provides everything necessary for quickly getting started with LLMs on OGA…”
   • [probe] “PROBE runtime (recorded 2026-09-15): the Ryzen AI Software documentation at ryzenai.docs.amd.com answered a keyless curl naming Ryzen AI — t…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K—none0/10

   The evidence pack contains no published TOPS figure for an NPU on the desktop 285K itself — the official spec page lists AI frameworks 'Supported by CPU' only (OpenVINO, WindowsML, DirectML, ONNX RT, WebNN), with no NPU TOPS number or dedicated NPU spec line, and generic oneAPI/VTune mentions of 'NPUs from Intel' are not tied to this specific SKU's hardware spec.

   • [claimed-docs] “AI Software Frameworks Supported by CPU OpenVINO™, WindowsML, DirectML, ONNX RT, WebNN”
   • [claimed-docs] “Optimize performance on client GPUs and NPUs from Intel with new analysis tool features in Intel® VTune™ Profiler.”
   • [claimed-docs] “Maximize AI PC inference capabilities from large language models (LLM) to image generation with Intel® oneAPI Deep Neural Network Library (o…”

   proof ↗⚑ flag

Runtime support

1. developerMainstream local-AI runtimes target this silicon — llama.cpp, MLX, ONNX Runtime, or the vendor's own AI software stack document support for its CPU, GPU, or NPU

   weight 2 · round to AMD Ryzen AI Max+ 395
   AMD Ryzen AI Max+ 395✓fullcommunity8/10

   AMD's own Ryzen AI docs confirm ONNX Runtime with Vitis AI Execution Provider support, and explicitly mention the Lemonade SDK enabling llama.cpp on this platform; community hands-on reports independently confirm llama.cpp (Vulkan) and Ollama running well on Strix Halo silicon. missing for 10: explicit MLX support (Apple-specific, not applicable here but leaves a gap in the story's named runtimes), and clearer first-party NPU-specific runtime benchmarks beyond GPU/CPU token-rate anecdotes.

   • [claimed-docs] “This allows developers to build and deploy models trained in PyTorch or TensorFlow and run them directly on laptops powered by Ryzen AI usin…”
   • [claimed-docs] “The AI model is deployed using the ONNX Runtime with either C++ or Python APIs.”
   • [claimed-docs] “the Lemonade SDK, which is multi-vendor open-source software that provides everything necessary for quickly getting started with LLMs on OGA…”
   • [community] “Own a framework desktop 128gb. GPU bandwidth limits the amount of tokens/sec that you get. I've mostly been running QWEN-3.6-35B-A3B at Q8 a…”
   • [community] “I'm running qwen3.6, gpt-oss and embeddinggemma with Ollama on Ryzen 7 8700G + 96 GB DDR5 with 12-14 tokens/sec... On Strix Halo it will run…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K✓fullclaimed7/10

   Intel's own spec page explicitly lists AI software frameworks supported by the CPU—OpenVINO, ONNX Runtime, DirectML, WindowsML, WebNN—and Intel's developer docs describe oneAPI/oneDNN and PyTorch optimizations tuned for Core Ultra processors and Arc GPUs, directly naming mainstream local-AI runtimes as supported. Missing for 10: explicit llama.cpp/MLX mention, and independent hands-on confirmation that these runtimes actually run well on the 285K specifically (reviews focus on gaming/productivity benchmarks, not AI runtime performance).

   • [claimed-docs] “AI Software Frameworks Supported by CPU OpenVINO™, WindowsML, DirectML, ONNX RT, WebNN”
   • [claimed-docs] “Maximize AI PC inference capabilities from large language models (LLM) to image generation with Intel® oneAPI Deep Neural Network Library (o…”

   proof ↗⚑ flag

Llm memory

1. ai-native userRun a large local LLM (70B-class, quantized) on this platform — enough addressable memory and published memory bandwidth to make local inference practical

   weight 3 · round to AMD Ryzen AI Max+ 395
   AMD Ryzen AI Max+ 395~partialcommunity6/10

   Community hands-on reports confirm the platform (128GB unified memory) can actually run large quantized models like Qwen3.5-122B-A10B at Q4 and similar 35B+ models via llama.cpp, showing real-world feasibility of 70B-class local inference. However, users consistently note memory bandwidth is the limiting factor for tokens/sec, and no official AMD-published memory bandwidth spec appears in the evidence pack. Missing for 10: an official AMD-published memory bandwidth figure, first-party benchmarks for 70B-class models, and confirmation that performance is 'practical' (not just possible) at long context lengths.

   • [community] “Own a framework desktop 128gb. GPU bandwidth limits the amount of tokens/sec that you get. I've mostly been running QWEN-3.6-35B-A3B at Q8 a…”
   • [community] “I'm running qwen3.6, gpt-oss and embeddinggemma with Ollama on Ryzen 7 8700G + 96 GB DDR5 with 12-14 tokens/sec... On Strix Halo it will run…”
   • [community] “With 128 GB of unified memory, large models can fit into memory, but memory bandwidth quickly becomes the limiting factor. With long context…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K—none0/10

   Evidence shows generic 'AI PC' and LLM-tooling marketing (oneDNN/PyTorch optimizations, OpenVINO support) but no published memory bandwidth figures, no addressable-memory/capacity specs, and no benchmarks or claims about running 70B-class quantized models; the CPU is a desktop chip relying on dual-channel DDR5 which is not evidenced as sufficient for practical 70B inference. Missing for 10: memory bandwidth specs, evidence of large-model (70B) local inference support, capacity/quantization guidance, and any hands-on corroboration of local large-LLM performance.

   • [claimed-docs] “Maximize AI PC inference capabilities from large language models (LLM) to image generation with Intel® oneAPI Deep Neural Network Library (o…”
   • [claimed-docs] “AI Software Frameworks Supported by CPU OpenVINO™, WindowsML, DirectML, ONNX RT, WebNN”

   proof ↗⚑ flag

Memory spec

1. developerSize memory-bound workloads from the vendor's own numbers — published memory type, capacity ceiling, and bandwidth (or spec detail complete enough to derive it)

   weight 2 · round to Intel Core Ultra 9 285K
   AMD Ryzen AI Max+ 395—none0/10

   The evidence pack includes AMD's spec page items but none cite memory type, capacity ceiling, or bandwidth figures (only TDP, NVMe, OS, and display specs are shown); community posts discuss bandwidth being a bottleneck qualitatively but no vendor numbers are cited to let a developer size workloads.

   • [claimed-docs] “AMD Configurable TDP (cTDP) 45-120W”
   • [claimed-docs] “Max Displays 4”
   • [community] “Own a framework desktop 128gb. GPU bandwidth limits the amount of tokens/sec that you get. I've mostly been running QWEN-3.6-35B-A3B at Q8 a…”
   • [community] “With 128 GB of unified memory, large models can fit into memory, but memory bandwidth quickly becomes the limiting factor. With long context…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K~partialprobed4/10

   Evidence shows ECC memory support and vPro details from Intel's ARK spec page, and community reports confirm a rated DDR5 speed (DDR5-6400) and CUDIMM support, but no explicit vendor-published memory capacity ceiling or bandwidth figures are quoted in the pack. The probe confirms ARK has a 'machine-fetchable spec sheet' including memory support, but the actual numbers aren't captured here. missing for 10: explicit max memory capacity (GB), number of channels, and bandwidth (GB/s) figures from vendor docs.

   • [claimed-docs] “ECC Memory Supported ‡ Yes”
   • [community] “When running with Corsair DDR5-8000 DIMMs, I was encountering compiler segmentation faults occasionally... With some DDR5-6000 DIMMs I also …”
   • [community] “Pros: Productivity performance, Power consumption and efficiency, Support for CUDIMM memory, Relaxed cooling requirements, Higher memory OC …”
   • [probe] “PROBE runtime (recorded 2026-09-15): Intel's ARK specifications page for the Core Ultra 9 285K answered a keyless curl and names the part — …”

   proof ↗⚑ flag

Platform io

1. developerThe platform has documented I/O headroom — PCIe generation and lanes, fast storage, and external connectivity specs I can plan a build or dock setup around

   weight 2 · round to AMD Ryzen AI Max+ 395
   AMD Ryzen AI Max+ 395~partialclaimed2/10

   AMD's product page lists some I/O-adjacent specs (NVMe boot/RAID support, max 4 displays, OS support) but there is no documented PCIe generation or lane count, and no external connectivity specs (USB/Thunderbolt/dock) to plan a build around. missing for 10: PCIe generation, PCIe lane count, USB/Thunderbolt/external port specs, dock compatibility documentation.

   • [claimed-docs] “NVMe Support Boot , RAID0 , RAID1”
   • [claimed-docs] “Max Displays 4”
   • [claimed-docs] “Windows 11 - 64-Bit Edition , RHEL x86 64-Bit , Ubuntu x86 64-Bit”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K—none0/10

   The evidence pack contains no specifics on PCIe generation/lane counts, storage interface support, or external connectivity (USB/Thunderbolt) specs for the 285K platform — only marketing copy, memory/vPro/ECC details, and Linux/Windows performance reports. This axis clearly applies to a desktop CPU/platform story, but no documentation of I/O headroom is present, so it cannot be credited. Missing for 10: PCIe lane/generation breakdown, storage (M.2/NVMe) specs, USB/Thunderbolt connectivity details, chipset I/O documentation.

   • [claimed-docs] “AI Software Frameworks Supported by CPU OpenVINO™, WindowsML, DirectML, ONNX RT, WebNN”
   • [claimed-docs] “ECC Memory Supported ‡ Yes”
   • [claimed-docs] “Intel vPro® Eligibility ‡ Intel vPro® Enterprise”

   proof ↗⚑ flag

Socket longevity

1. power-userUpgrade the CPU without replacing the platform — a documented socket with a stated multi-generation support commitment

   weight 2 · round drawn
   AMD Ryzen AI Max+ 395—none0/10

   The evidence pack contains no mention of a socket, upgrade path, or multi-generation platform commitment for the Ryzen AI Max+ 395; all specs shown are TDP, display, storage, and OS support with no socket/upgrade language. Community citations also reference it in fixed-configuration devices (e.g., Framework Desktop, mini-PCs) rather than a socketed upgrade path. missing for 10: any documented socket name, generational upgrade commitment, or evidence of user-replaceable CPU in a motherboard.

   • [claimed-docs] “AMD Configurable TDP (cTDP) 45-120W”
   • [claimed-docs] “Max Displays 4”
   • [community] “Own a framework desktop 128gb. GPU bandwidth limits the amount of tokens/sec that you get. I've mostly been running QWEN-3.6-35B-A3B at Q8 a…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K—none0/10

   The evidence pack contains no mention of the CPU's socket (LGA1851) or any stated multi-generation upgrade commitment; specs pages cover cache/clocks/vPro/ECC but not platform longevity. This axis clearly applies to a desktop CPU purchase decision, but no documentation here confirms or denies a multi-gen socket promise.

   ⚑ flag

Mobile endurance

1. power-userThis chip powers thin, quiet, all-day-battery machines — shipping in fanless or low-power designs with credible battery-life evidence

   weight 2 · round drawn
   AMD Ryzen AI Max+ 395—none0/10

   The evidence pack shows a configurable TDP range of 45–120W and OS/platform specs, but contains no mention of fanless designs, thin-and-light chassis, or battery-life benchmarks; a community comment even complains AMD hasn't focused on reducing power usage. This is a fair axis for a laptop-class chip, but no supporting evidence exists.

   • [claimed-docs] “AMD Configurable TDP (cTDP) 45-120W”
   • [community] “The performance is incredible... Now when will AMD put some real effort in to reducing power usage?”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K—none0/10

   The axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na". (na/none harmonized at arena bring-up — see pipeline/scripts/na-harmonize.ts.)

   ⚑ flag

Perf per watt

1. creatorLong renders and exports don't throttle away — documented power envelopes and independent testing showing strong sustained performance per watt

   weight 3 · round to Intel Core Ultra 9 285K
   AMD Ryzen AI Max+ 395~partialcommunity3/10

   AMD documents a configurable TDP range (45-120W) for the chip, but there is no independent benchmark or hands-on testing demonstrating sustained performance-per-watt or absence of thermal throttling during long renders/exports; community comments focus on LLM token throughput and even note frustration that AMD hasn't addressed power usage. missing for 10: independent sustained-load/thermal throttling benchmarks, creator-workload (render/export) power efficiency data, cooling/chassis-specific real-world tests.

   • [claimed-docs] “AMD Configurable TDP (cTDP) 45-120W”
   • [community] “The performance is incredible... Now when will AMD put some real effort in to reducing power usage?”
   • [community] “Terrible benchmarks. They should probably compare the 395 GPU against the 9950 CPU or against a 7600/9060. If you're just going to use the C…”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K~partialcommunity5/10

   Independent reviews (Tom's Hardware) confirm the 285K delivers strong productivity/rendering gains with improved power consumption and efficiency versus its 14900K predecessor, and Intel's spec pages document power envelopes (base/turbo power). However, the same reviews note it trails AMD's 9950X in raw perf-per-watt, and no evidence specifically tests sustained long-render/export thermal throttling behavior. Missing for 10: dedicated sustained-load/throttling benchmarks over long export durations, explicit perf-per-watt superiority claims corroborated independently, and creator-specific workload power testing.

   • [community] “Intel's Core Ultra 9 285K makes strong gains in productivity workloads, but it struggles to match its prior-gen counterpart in gaming perfor…”
   • [community] “Pros: Productivity performance, Power consumption and efficiency, Support for CUDIMM memory, Relaxed cooling requirements, Higher memory OC …”
   • [community] “Compared to its predecessor, 14900K, it ends up both faster (barring some outliers) and significantly more power efficient. Compared to Zen …”
   • [claimed-docs] “ECC Memory Supported ‡ Yes”

   proof ↗⚑ flag

Spec disclosure

1. power-userComparison-shop from a real spec sheet — the vendor publishes clocks, power, memory support, and AI TOPS with test conditions, instead of marketing adjectives

   weight 2 · round to Intel Core Ultra 9 285K
   AMD Ryzen AI Max+ 395~partialcommunity5/10

   AMD's product page does publish concrete spec-sheet items (cTDP 45-120W, NVMe RAID support, OS support, display counts, voltage offset support) rather than pure marketing prose, and the Ryzen AI docs give technical detail on the software stack. However, the pack contains no explicit clock-speed figures or AI TOPS number with stated test conditions, and community commentary explicitly complains that AMD's benchmark comparisons are misleading/incomplete rather than rigorously documented. missing for 10: published boost/base clock figures, an explicit AI TOPS figure with test-methodology footnotes, and independent verification that the spec sheet's numbers hold up under real-world testing.

   • [claimed-docs] “AMD Configurable TDP (cTDP) 45-120W”
   • [claimed-docs] “Curve Optimizer Voltage Offsets Yes”
   • [claimed-docs] “NVMe Support Boot , RAID0 , RAID1”
   • [claimed-docs] “Windows 11 - 64-Bit Edition , RHEL x86 64-Bit , Ubuntu x86 64-Bit”
   • [claimed-docs] “Max Displays 4”
   • [community] “Terrible benchmarks. They should probably compare the 395 GPU against the 9950 CPU or against a 7600/9060. If you're just going to use the C…”
   • [community] “It would have been interesting to see the graphics performance and AI performance of the 395 compared to discrete gpus... how does it stack …”

   proof ↗⚑ flag

   Intel Core Ultra 9 285K~partialprobed6/10

   Intel's ARK specifications page (and the runtime probe) confirms a machine-fetchable spec sheet listing clocks, cache, ECC support, vPro eligibility, and supported AI frameworks (OpenVINO, DirectML, ONNX RT), giving power-users concrete comparable data beyond marketing language. However, the evidence pack lacks explicit AI TOPS figures with stated test conditions, detailed base/turbo power figures, or memory speed/latency specs with test methodology, and much of the docs pack is marketing copy rather than spec data. missing for 10: explicit AI TOPS numbers with test conditions, full power/TDP breakdown, independent verification of spec accuracy.

   • [claimed-docs] “AI Software Frameworks Supported by CPU OpenVINO™, WindowsML, DirectML, ONNX RT, WebNN”
   • [claimed-docs] “ECC Memory Supported ‡ Yes”
   • [claimed-docs] “Intel vPro® Eligibility ‡ Intel vPro® Enterprise”
   • [probe] “PROBE runtime (recorded 2026-09-15): Intel's ARK specifications page for the Core Ultra 9 285K answered a keyless curl and names the part — …”

   proof ↗⚑ flag