AMD Ryzen AI Max+ 395 vs AMD Ryzen 9 9950X3D
AMD Ryzen AI Max+ 395 wins · 9–6 (5 drawn)
Agenticness — how well agents can access and operate the productAgenticness
How well agents can access and operate the product
Agent access
ai-native userPoint an agent at llms.txt or agent-oriented docs
weight 2 · round to AMD Ryzen AI Max+ 395A 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…”
ai-native userRun the product headlessly / in CI for automation
weight 2 · round to AMD Ryzen AI Max+ 395The 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”
ai-native userDrive the product through a documented public API
weight 3 · round to AMD Ryzen AI Max+ 395AMD 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…”
ai-native userBuild against official SDKs
weight 2 · round to AMD Ryzen AI Max+ 395AMD 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…”
Agentic features
ai-native userOperate the product with natural-language commands
weight 2 · round drawnAMD Ryzen AI Max+ 395none0/10The 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.)
Compute performance — stories about compute performance in this arenaCompute performance
Stories about compute performance in this arena
Heavy compute
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 AMD Ryzen 9 9950X3DAMD Ryzen AI Max+ 395none0/10No 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.
AMD docs confirm core/thread count (up to 32 threads) and boost tuning via PBO/Ryzen Master, and independent reviews (TechPowerup, Tom's Hardware) corroborate strong productivity performance alongside class-leading gaming benchmarks. However, evidence lacks explicit compile-time or multi-threaded workload benchmarks (e.g., Cinebench multi-core, code build times) to directly validate 'compile big codebases' claims. Missing for 10: explicit multi-core/compile benchmark numbers, detailed core/thread topology breakdown, independent parallel-workload testing beyond general productivity mentions.
- [claimed-docs] “premium capabilities like time‑saving PCIe® 5.0 storage support, ultra‑fast Wi‑Fi® 6E3, AMD EXPO™ technology4, up to 32 processing threads”
- [claimed-docs] “Easily personalize performance by tweaking and tuning your AMD Ryzen processor with one click Precision Boost Overdrive or manually through …”
- [community] “Review verdict: 9950X3D is the fastest 16-core chip in gaming, achieves near-parity gaming performance with the 9800X3D, retains strong prod…”
- [community] “Reviewer's Pros/Cons list for the 9950X3D: Fastest 16-core chip in gaming, near-parity gaming with the 9800X3D, strong productivity performa…”
Single thread
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 AMD Ryzen 9 9950X3DAMD Ryzen AI Max+ 395none0/10No 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…”
Independent benchmarks (TechPowerUp, Tom's Hardware) confirm the 9950X3D delivers leading real-world single-thread-sensitive performance (gaming, near-parity with 9800X3D, beating Intel chips by 26-37%), and community commentary corroborates responsiveness gains beyond marketing claims. Missing for 10: dedicated single-thread-only benchmark isolation (e.g. Cinebench single-core scores) rather than aggregate gaming/productivity figures.
- [community] “Review verdict: 9950X3D is the fastest 16-core chip in gaming, achieves near-parity gaming performance with the 9800X3D, retains strong prod…”
- [community] “Benchmarks show the 9950X3D is 37% faster than Intel's Core 9 285K on average in 1080p gaming and 26% faster than the Core i9-14900K, while …”
- [community] “Reviewer's Pros/Cons list for the 9950X3D: Fastest 16-core chip in gaming, near-parity gaming with the 9800X3D, strong productivity performa…”
- [community] “Commenter surprised that even non-gaming tasks benefit significantly from the 3D V-Cache on the 9950X3D, speculating this differs from prior…”
Dev experience — day-to-day developer experience — setup friction, docs, debugging, iteration speedDev experience
Day-to-day developer experience — setup friction, docs, debugging, iteration speed
Toolchain
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 to AMD Ryzen AI Max+ 395AMD 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…”
Evidence shows official OS compatibility listings (RHEL, Ubuntu) and community-confirmed AVX-512 support, suggesting baseline OS/tooling recognition, but there is no evidence of AMD-specific compiler optimization guides, toolchain documentation, or first-class developer tooling status. Missing for 10: official compiler/optimization guidance (GCC/LLVM tuning docs), developer-targeted architecture manuals, and independent confirmation of tooling maturity beyond basic OS support.
- [claimed-docs] “RHEL x86 64-Bit , Ubuntu x86 64-Bit”
- [community] “Review verdict: 9950X3D is the fastest 16-core chip in gaming, achieves near-parity gaming performance with the 9800X3D, retains strong prod…”
- [community] “Reviewer's Pros/Cons list for the 9950X3D: Fastest 16-core chip in gaming, near-parity gaming with the 9800X3D, strong productivity performa…”
Virtualization
developerVMs and containers run well on this silicon — documented virtualization support and mainstream hypervisor and Docker workflows
weight 2 · round drawnAMD Ryzen AI Max+ 395none0/10The 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”
AMD Ryzen 9 9950X3Dnone0/10Evidence pack shows OS support listing (RHEL, Ubuntu) but no documentation or hands-on evidence about virtualization features (SVM/AMD-V, IOMMU), hypervisor compatibility (KVM, VMware, Hyper-V), or Docker/container workflows on this CPU. missing for 10: virtualization extension docs, hypervisor compatibility evidence, container/Docker workflow evidence, developer hands-on reports.
- [claimed-docs] “RHEL x86 64-Bit , Ubuntu x86 64-Bit”
Gaming media — stories about gaming media in this arenaGaming media
Stories about gaming media in this arena
Gaming fps
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 AMD Ryzen 9 9950X3DAMD Ryzen AI Max+ 395none0/10The 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 …”
Vendor claims of massive 3D V-Cache (up to 208MB) driving 'ultimate gaming performance' are corroborated by independent benchmarks showing the 9950X3D is the fastest 16-core gaming chip, beats Intel's 285K by 37% and 14900K by 26%, and nearly matches the dedicated gaming champion 9800X3D, with community discussion confirming real-world benefits from the cache. Missing for 10: no integrated GPU gaming benchmarks or iGPU class comparison since this is a desktop chip without significant iGPU gaming evidence.
- [claimed-docs] “AMD combines its flagship Ryzen 9000X3D Series processors with up to a colossal 208MB of on-chip memory, paired with the most advanced proce…”
- [claimed-docs] “AMD 3D V-Cache technology delivers a huge game performance advantage with up to 208MB of on-chip memory, available on AMD Ryzen X3D Series p…”
- [community] “Review verdict: 9950X3D is the fastest 16-core chip in gaming, achieves near-parity gaming performance with the 9800X3D, retains strong prod…”
- [community] “Benchmarks show the 9950X3D is 37% faster than Intel's Core 9 285K on average in 1080p gaming and 26% faster than the Core i9-14900K, while …”
- [community] “Reviewer's Pros/Cons list for the 9950X3D: Fastest 16-core chip in gaming, near-parity gaming with the 9800X3D, strong productivity performa…”
- [community] “Commenter surprised that even non-gaming tasks benefit significantly from the 3D V-Cache on the 9950X3D, speculating this differs from prior…”
Media engines
creatorHardware media engines carry my editing and streaming — documented hardware encode and decode (AV1, HEVC, ProRes-class) on the chip itself
weight 2 · round drawnAMD Ryzen AI Max+ 395none0/10Evidence 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.
Local ai — stories about local ai in this arenaLocal ai
Stories about local ai in this arena
Npu sdk
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+ 395Strong 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…”
AMD Ryzen 9 9950X3Dnone0/10The evidence only shows generic marketing links to 'Ryzen AI Software For Developers' but never states that the 9950X3D itself contains an NPU, nor gives any TOPS figure or precision spec, nor confirms an SDK/runtime ships for this specific chip (Ryzen AI is typically a mobile-chip NPU feature). Missing for 10: NPU presence confirmation on this SKU, a published TOPS figure with precision, and evidence of a shipping SDK/runtime tied to this desktop chip.
- [claimed-docs] “Learn About Ryzen AI Software For Developers”
Runtime support
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+ 395AMD'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…”
AMD Ryzen 9 9950X3Dnone0/10The evidence only shows a generic marketing link to 'Ryzen AI Software For Developers' with no indication it applies to the 9950X3D desktop CPU (which lacks an NPU) and no mention of llama.cpp, MLX, ONNX Runtime, or any concrete runtime documentation naming this chip. Linux OS support (RHEL/Ubuntu) is noted but that's generic OS compatibility, not local-AI runtime support documentation.
- [claimed-docs] “Learn About Ryzen AI Software For Developers”
- [claimed-docs] “RHEL x86 64-Bit , Ubuntu x86 64-Bit”
Memory io — stories about memory io in this arenaMemory io
Stories about memory io in this arena
Llm memory
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+ 395Community 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…”
AMD Ryzen 9 9950X3Dnone0/10The evidence pack contains no data on memory bandwidth, maximum supported RAM capacity, or any benchmarks/claims about running large (70B-class) LLMs locally; it focuses on gaming, overclocking, and 3D V-Cache for gaming/productivity workloads. Missing for 10: published memory bandwidth figures, max RAM/channel config data, and any LLM inference benchmarks or vendor guidance for local AI workloads.
Memory spec
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 drawnAMD Ryzen AI Max+ 395none0/10The 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…”
AMD Ryzen 9 9950X3Dnone0/10The evidence pack contains no published memory type (e.g., DDR5 speed), no memory capacity ceiling, and no bandwidth figures or clock/channel specs from which bandwidth could be derived — only marketing mentions of EXPO overclocking, ECC 'requires mobo support', and on-die 3D V-Cache (which is not system RAM). Missing for 10: memory type/speed spec, max supported memory capacity, number of channels, and bandwidth or derivable throughput numbers.
- [claimed-docs] “AMD EXPO™ one-touch memory overclocking”
- [claimed-docs] “ECC Support Yes (Requires mobo support)”
- [claimed-docs] “AMD 3D V-Cache technology delivers a huge game performance advantage with up to 208MB of on-chip memory, available on AMD Ryzen X3D Series p…”
Platform upgrade — stories about platform upgrade in this arenaPlatform upgrade
Stories about platform upgrade in this arena
Platform io
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 9 9950X3DAMD'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”
AMD's docs confirm PCIe 5.0 storage support, NVMe RAID/boot configurations, and ECC memory support (with motherboard dependency), giving some concrete I/O planning data. However, missing for 10: exact PCIe lane count/allocation, USB/Thunderbolt external connectivity specs, and dock-compatible port bandwidth details that a developer would need for a full build-planning picture.
- [claimed-docs] “premium capabilities like time‑saving PCIe® 5.0 storage support, ultra‑fast Wi‑Fi® 6E3, AMD EXPO™ technology4, up to 32 processing threads”
- [claimed-docs] “NVMe Support Boot , RAID0 , RAID1 , RAID5 , RAID10”
- [claimed-docs] “ECC Support Yes (Requires mobo support)”
Socket longevity
power-userUpgrade the CPU without replacing the platform — a documented socket with a stated multi-generation support commitment
weight 2 · round to AMD Ryzen 9 9950X3DAMD Ryzen AI Max+ 395none0/10The 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…”
AMD explicitly documents a multi-year socket support commitment ('AMD is the only processor manufacturer committed to multi-year socket support... start with what fits your build today and upgrade to next-gen performance tomorrow'), directly matching the platform-upgrade story. Missing for 10: independent/community confirmation of actual multi-generation compatibility (e.g. AM5 supporting multiple CPU generations in practice) and specifics on how many generations/years are guaranteed.
- [claimed-docs] “AMD is the only processor manufacturer committed to multi-year socket support, giving you the freedom to start with what fits your build tod…”
- [claimed-docs] “Invest in a platform that grows with your needs over time...you can start with what fits your build today and upgrade to next-gen performanc…”
Power efficiency — stories about power efficiency in this arenaPower efficiency
Stories about power efficiency in this arena
Mobile endurance
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 drawnAMD Ryzen AI Max+ 395none0/10The 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?”
Perf per watt
creatorLong renders and exports don't throttle away — documented power envelopes and independent testing showing strong sustained performance per watt
weight 3 · round to AMD Ryzen 9 9950X3DAMD 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…”
AMD's marketing highlights creator workloads (DaVinci Resolve, 3D rendering) and tunable power via PBO, and independent reviews broadly call the chip 'energy efficient' with 'strong productivity performance,' but none of the evidence gives documented power envelope specs or dedicated long-render/export sustained perf-per-watt testing. missing for 10: explicit TDP/power envelope documentation, independent sustained-load (render/export) thermal-throttle or perf-per-watt benchmarks.
- [claimed-docs] “Easily personalize performance by tweaking and tuning your AMD Ryzen processor with one click Precision Boost Overdrive or manually through …”
- [claimed-docs] “Our DaVinci Resolve editors, colorists, and visual effects artists working in the film and television need all the processing power they can…”
- [community] “Review verdict: 9950X3D is the fastest 16-core chip in gaming, achieves near-parity gaming performance with the 9800X3D, retains strong prod…”
- [community] “Reviewer's Pros/Cons list for the 9950X3D: Fastest 16-core chip in gaming, near-parity gaming with the 9800X3D, strong productivity performa…”
Spec transparency — stories about spec transparency in this arenaSpec transparency
Stories about spec transparency in this arena
Spec disclosure
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 AMD Ryzen AI Max+ 395AMD'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 …”
AMD Ryzen 9 9950X3Dnone0/10The evidence pack is dominated by marketing language ('personalize performance', 'colossal 208MB', 'accelerate every step') rather than a transparent spec sheet; while a few technical bullets exist (ECC support, NVMe RAID modes, supported OSes), there is no evidence of published clock speeds, TDP/power figures, memory speed/type support, or AI TOPS numbers with stated test conditions.
- [claimed-docs] “ECC Support Yes (Requires mobo support)”
- [claimed-docs] “NVMe Support Boot , RAID0 , RAID1 , RAID5 , RAID10”
- [claimed-docs] “RHEL x86 64-Bit , Ubuntu x86 64-Bit”
- [claimed-docs] “AMD combines its flagship Ryzen 9000X3D Series processors with up to a colossal 208MB of on-chip memory, paired with the most advanced proce…”
- [claimed-docs] “AMD 3D V-Cache technology delivers a huge game performance advantage with up to 208MB of on-chip memory, available on AMD Ryzen X3D Series p…”
Not comparable on these axes
ai-native userPlug MCP servers into this product so it can use their tools
weight 3 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware CPU/APU product, not an agent or platform with an MCP client/server role; plugging MCP servers into a chip is a category error.
ai-native userConnect an agent via an official MCP server
weight 3 · not comparableAMD Ryzen AI Max+ 395n/aAMD Ryzen AI Max+ 395 is a hardware CPU/APU product, not an agent or service that would expose an MCP server; this axis is a category error for a processor.
ai-native userUse an official CLI
weight 2 · not comparableAMD Ryzen AI Max+ 395none0/10Evidence describes SDKs (ONNX Runtime, Quark, Lemonade SDK) and APIs but never mentions an official command-line interface for Ryzen AI Max+ 395 developers; no CLI tool, install command, or CLI documentation is cited.
ai-native userIssue scoped/least-privilege API credentials for an agent
weight 2 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware CPU/SoC product, not an API service or IAM system; issuing scoped API credentials for agents is a category error for this product type.
ai-native userSubscribe to events via webhooks
weight 2 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware CPU/APU product; webhooks/event subscriptions are a software/service integration concern that does not apply to a silicon chip's own capabilities.
ai-native userGet AI-generated insights and suggestions from my data inside the product
weight 2 · not comparableAMD Ryzen AI Max+ 395n/aAMD Ryzen AI Max+ 395 is a hardware processor/SDK platform for running AI models, not an end-user application that holds 'my data' and surfaces insights from it; this story targets data-centric SaaS/app products, so the axis is a category mismatch for a chip/SDK.
AMD Ryzen 9 9950X3Dn/aThis is a hardware CPU product; 'AI-generated insights from data inside the product' is a software/application-layer capability, not something a processor itself delivers. The evidence only covers performance, overclocking, and benchmarks, confirming this axis is a category error for a CPU.
ai-native userSet up automations that run autonomously in the background
weight 2 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware CPU/APU product, not an automation/orchestration platform; setting up autonomous background automations is a software/agent-platform axis that doesn't apply to a chip.
ai-native userDelegate tasks to a built-in AI assistant inside the product
weight 3 · not comparableAMD Ryzen AI Max+ 395n/aAMD Ryzen AI Max+ 395 is a hardware processor/SoC with an AI development SDK for building and running models; it is not a product with a built-in end-user AI assistant to delegate tasks to. This axis applies to consumer assistant products/agents, not to a CPU/NPU platform whose evidence only covers developer tooling like ONNX Runtime, Quark quantization, and Lemonade SDK.
ai-native userExplore an interactive API reference with runnable examples
weight 2 · not comparableAMD Ryzen AI Max+ 395none0/10The evidence shows static documentation pages (docs-1 through docs-4) and probes explicitly confirming no OpenAPI/swagger spec and no interactive API reference (probe-3 shows 404s for all candidate OpenAPI paths). There is no evidence of runnable examples or an interactive API explorer anywhere in the pack.
- [probe] “PROBE openapi: all candidate paths 404 (https://ryzenai.docs.amd.com/openapi.json, https://ryzenai.docs.amd.com/swagger.json, https://ryzena…”
- [probe] “PROBE docs-md: HTTP 404 at https://ryzenai.docs.amd.com/en/latest/.md”
- [claimed-docs] “The AI model is deployed using the ONNX Runtime with either C++ or Python APIs.”
ai-native userDownload a machine-readable API spec (OpenAPI or equivalent)
weight 2 · not comparableAMD Ryzen AI Max+ 395none0/10AMD's Ryzen AI docs describe C++/Python SDK APIs (ONNX Runtime, Vitis AI EP) but no machine-readable OpenAPI/Swagger spec was found; explicit probes to /openapi.json, /swagger.json, and similar paths all returned 404.
- [probe] “PROBE openapi: all candidate paths 404 (https://ryzenai.docs.amd.com/openapi.json, https://ryzenai.docs.amd.com/swagger.json, https://ryzena…”
- [claimed-docs] “The AI model is deployed using the ONNX Runtime with either C++ or Python APIs.”
ai-native userTest against a sandbox environment without touching production data
weight 1 · not comparableAMD Ryzen AI Max+ 395n/aThis story concerns sandboxed testing environments isolated from production data — a software/platform deployment concept that doesn't apply to a hardware CPU/APU product line like this one.
ai-native userRely on versioned APIs with a documented deprecation policy
weight 2 · not comparableAMD Ryzen AI Max+ 395none0/10Evidence shows the Ryzen AI software stack exposes C++/Python APIs via ONNX Runtime and mentions SDKs like Quark and Lemonade, but nothing documents API versioning practices or a deprecation policy for these interfaces. Missing for 10: any versioning scheme, changelog, or explicit deprecation/EOL policy for the Ryzen AI APIs.
- [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…”
ai-native userPerform bulk operations across many items at once
weight 2 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware CPU/APU product; 'bulk operations across many items' is a software/application-level automation feature not applicable to a silicon chip's product story.
ai-native userDefine rules that trigger actions automatically on events
weight 3 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware CPU/SoC product; rule-based automation triggers on events is an application/software-platform feature, not something a processor exposes. The story is a category error for this product type.
ai-native userSchedule recurring jobs or workflows
weight 2 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware CPU/APU product, not a workflow/job orchestration platform; scheduling recurring jobs is a software/OS-level concern outside this product's category.
ai-native userVersion, review, and roll back my automations
weight 1 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware CPU/SoC product; versioning, reviewing, and rolling back 'automations' is a software/workflow-tool concept that does not apply to a chip's category.
ai-native userDo everything through the API that I can do in the UI
weight 2 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware processor product, not a UI/application with an API vs UI parity question — there is no 'UI' for a CPU to compare against an API. The axis is a category error for a hardware SKU.
ai-native userExport all of my data in open formats and leave
weight 3 · not comparableAMD Ryzen AI Max+ 395n/aAMD Ryzen AI Max+ 395 is a hardware processor/chip, not a data-holding service or platform where a user stores personal data that could be exported; the 'export data and leave' story is a category error for this product type.
ai-native userRead the product's source under an open license
weight 2 · not comparableAMD Ryzen AI Max+ 395none0/10The product is a proprietary AMD CPU/APU design; while some accompanying SDKs (Lemonade, ONNX EP) are noted as open-source, there is no evidence that the chip's own source/design (microarchitecture, RTL, etc.) is available under any open license. The axis is fair to ask (some hardware vendors do open designs) but no evidence supports it here.
- [claimed-docs] “the Lemonade SDK, which is multi-vendor open-source software that provides everything necessary for quickly getting started with LLMs on OGA…”
ai-native userSelf-host the core product
weight 3 · not comparableAMD Ryzen AI Max+ 395n/aThe Ryzen AI Max+ 395 is a physical CPU/APU, not a software service or platform with a hosted vs. self-hosted deployment choice — 'self-hosting the core product' is a category error for silicon hardware, which is inherently run on the owner's own machine by nature rather than as a deployment option.
ai-native userChoose where my data is stored (region/residency)
weight 2 · not comparableAMD Ryzen AI Max+ 395n/aAMD Ryzen AI Max+ 395 is a hardware CPU/APU chip; data residency/region storage is a cloud-service/SaaS concept and does not apply to a local processor product.
ai-native userPrevent my data from being used to train AI models
weight 3 · not comparableAMD Ryzen AI Max+ 395none0/10The evidence describes local on-device AI inference (NPU, ONNX Runtime, local LLM execution via llama.cpp/Ollama) but contains no explicit statement about data-training opt-outs or privacy guarantees regarding model training; while local processing implies data doesn't leave the device, no documentation or claim addresses this story directly.
ai-native userControl data retention and deletion
weight 2 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware chip (CPU/APU), not a data-hosting service or SaaS product; data retention and deletion controls are not applicable to a physical processor — inference runs locally on-device and there is no vendor-hosted data lifecycle to manage.
ai-native userOpt out of telemetry and usage tracking
weight 2 · not comparableAMD Ryzen AI Max+ 395n/aThis is a hardware CPU/APU product; telemetry opt-out is a software/service privacy axis that doesn't apply to a physical processor SKU itself.