GeForce RTX 5090 vs Radeon RX 9070 XT
GeForce RTX 5090
NVIDIA Corporation
Radeon RX 9070 XT wins · 7–9 (11 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 Radeon RX 9070 XTNVIDIA's developer portal serves a valid llms.txt (HTTP 200, descriptive content) that an agent could use to discover CUDA/GPU-related docs, but this is at the general developer.nvidia.com domain rather than an RTX-5090-specific surface, and adjacent agent-friendly affordances (markdown doc exports, OpenAPI spec) are absent (404s). missing for 10: RTX-5090-specific llms.txt or agent doc entry point, working markdown/API exports, independent confirmation agents actually use this pathway successfully.
- [probe] “PROBE llms.txt: HTTP 200 at https://developer.nvidia.com/llms.txt # NVIDIA Developer > Comprehensive developer portal for NVIDIA accelerate…”
- [probe] “PROBE docs-md: HTTP 404 at https://developer.nvidia.com/cuda-toolkit.md”
- [probe] “PROBE openapi: all candidate paths 404 (https://developer.nvidia.com/openapi.json, https://developer.nvidia.com/swagger.json, https://develo…”
A direct probe confirms rocm.docs.amd.com/llms.txt returns HTTP 200 with structured content pointing to sub-project docs, giving an agent a genuine machine-readable entry point into the Radeon/ROCm software docs relevant to this GPU. Missing for 10: no evidence of llms.txt coverage at finer granularity (e.g., per-page) and no OpenAPI/agent API surface (which 404s), so agentic doc access is present but not comprehensive.
- [probe] “PROBE llms.txt: HTTP 200 at https://rocm.docs.amd.com/llms.txt # ROCm documentation > Note: ROCm documentation is split across multiple pro…”
- [probe] “PROBE openapi: all candidate paths 404 (https://rocm.docs.amd.com/openapi.json, https://rocm.docs.amd.com/swagger.json, https://rocm.docs.am…”
- [claimed-docs] “PyTorch on Windows updated with ROCm 7.2.1 on AMD Radeon graphics products and AMD Ryzen AI processors.”
ai-native userRun the product headlessly / in CI for automation
weight 2 · round to Radeon RX 9070 XTGeForce RTX 5090none0/10The 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.)
Evidence shows ROCm/Linux support with PyTorch, vLLM, and Llama.cpp for compute workloads (amd-rx-9070-xt-docs-2, -10, -11), which implies the GPU can be used for automated inference/training pipelines without a display, but there is no explicit documentation of headless operation, Docker/CI runner support, or automation-specific tooling. Missing for 10: explicit headless-mode docs, CI/container integration guides, and any hands-on evidence of running in automated pipelines.
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Linux® PyTorch, TensorFlow, JAX, ONNX”
- [claimed-docs] “**vLLM**: Full support.”
- [claimed-docs] “**Llama.cpp**: Supported for efficient inference.”
- [claimed-docs] “OS Support Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit”
ai-native userUse an official CLI
weight 2 · round drawnGeForce RTX 5090none0/10The 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.)
ai-native userDrive the product through a documented public API
weight 3 · round to Radeon RX 9070 XTCUDA Toolkit documentation describes a programmatic API (CUDA, cuTile Python/C++) for building GPU-accelerated applications, which is the closest analog to a 'documented public API' for this hardware product, but this is a low-level compute-kernel API, not an agent-drivable control interface, and direct probes found no OpenAPI/Swagger spec or machine-readable docs (404s). Missing for 10: a structured/machine-readable API spec (OpenAPI/REST), any agentic control surface, and independent corroboration of AI-native programmatic access beyond raw CUDA kernel programming.
- [claimed-docs] “The NVIDIA® CUDA® Toolkit provides a development environment for creating high-performance, GPU-accelerated applications.”
- [claimed-docs] “cuTile Python is an expression of the CUDA Tile programming model in Python.”
- [probe] “PROBE docs-md: HTTP 404 at https://developer.nvidia.com/cuda-toolkit.md”
- [probe] “PROBE openapi: all candidate paths 404 (https://developer.nvidia.com/openapi.json, https://developer.nvidia.com/swagger.json, https://develo…”
AMD documents a public software stack (ROCm) with APIs/libraries (HIP, PyTorch/TensorFlow/JAX/ONNX/vLLM/llama.cpp integration) that let developers programmatically drive the GPU for AI workloads, and llms.txt is served for documentation discovery. However, there is no dedicated machine-callable REST/OpenAPI interface — the openapi probe returned 404 on all candidate paths — so an AI agent cannot invoke a structured public API directly, only use ROCm's compiled libraries/frameworks. Missing for 10: a documented REST/OpenAPI or similarly agent-consumable API endpoint, independent confirmation of programmatic control beyond framework bindings.
- [claimed-docs] “PyTorch on Windows updated with ROCm 7.2.1 on AMD Radeon graphics products and AMD Ryzen AI processors.”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Linux® PyTorch, TensorFlow, JAX, ONNX”
- [claimed-docs] “**vLLM**: Full support.”
- [claimed-docs] “**Llama.cpp**: Supported for efficient inference.”
- [probe] “PROBE llms.txt: HTTP 200 at https://rocm.docs.amd.com/llms.txt # ROCm documentation > Note: ROCm documentation is split across multiple pro…”
- [probe] “PROBE openapi: all candidate paths 404 (https://rocm.docs.amd.com/openapi.json, https://rocm.docs.amd.com/swagger.json, https://rocm.docs.am…”
ai-native userBuild against official SDKs
weight 2 · round drawnNVIDIA provides official SDKs (CUDA Toolkit, CUDA Tile C++/cuTile Python, RTX Neural Shaders SDK, RTX Mega Geometry, Nsight tools) that developers can build AI/graphics applications against, all documented on NVIDIA's developer portal. Missing for 10: independent/hands-on developer corroboration of building against these SDKs, and deeper API reference/sample documentation beyond marketing-style descriptions.
- [claimed-docs] “The NVIDIA® CUDA® Toolkit provides a development environment for creating high-performance, GPU-accelerated applications.”
- [claimed-docs] “CUDA Tile C++ is an expression of the CUDA Tile programming model in C++. It's built on top of the CUDA Tile IR specification and allows you…”
- [claimed-docs] “cuTile Python is an expression of the CUDA Tile programming model in Python.”
- [claimed-docs] “The RTX Neural Shaders SDK lets developers train shader data on an RTX PRO workstation and accelerate neural representations with NVIDIA Ten…”
- [claimed-docs] “RTX Mega Geometry dramatically increases the geometric detail possible in ray-traced scenes, accelerating BVH building to enable up to 100x …”
- [claimed-docs] “NVIDIA Nsight Compute and Nsight System suite of tools designed to help developers optimize and increase performance of their applications.”
AMD provides official ROCm SDK documentation for the RX 9070 XT enabling development with PyTorch, TensorFlow, JAX, ONNX, vLLM, and llama.cpp on both Windows and Linux, with an llms.txt confirming machine-readable docs availability. Missing for 10: independent hands-on developer confirmation of SDK stability/completeness, and no evidence of API/openapi endpoints for programmatic integration.
- [claimed-docs] “PyTorch on Windows updated with ROCm 7.2.1 on AMD Radeon graphics products and AMD Ryzen AI processors.”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Linux® PyTorch, TensorFlow, JAX, ONNX”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Windows® PyTorch”
- [claimed-docs] “**vLLM**: Full support.”
- [claimed-docs] “**Llama.cpp**: Supported for efficient inference.”
- [claimed-docs] “The same ROCm™ stack that powers your desktop development on RDNA™ architecture GPUs also supports AMD Instinct™ accelerators on CDNA™ archi…”
- [probe] “PROBE llms.txt: HTTP 200 at https://rocm.docs.amd.com/llms.txt # ROCm documentation > Note: ROCm documentation is split across multiple pro…”
Agentic features
ai-native userGet AI-generated insights and suggestions from my data inside the product
weight 2 · round drawnGeForce RTX 5090none0/10The 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.)
ai-native userDelegate tasks to a built-in AI assistant inside the product
weight 3 · round drawnGeForce RTX 5090none0/10The 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.)
ai-native userOperate the product with natural-language commands
weight 2 · round drawnGeForce RTX 5090none0/10The 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.)
Ai compute — stories about ai compute in this arenaAi compute
Stories about ai compute in this arena
Inference stack
ai-native userThis GPU has a documented LLM inference story — low-precision formats (FP8/FP4) and supported serving stacks (TensorRT-LLM, vLLM, ROCm, llama.cpp) for this part
weight 3 · round drawnNVIDIA's RTX documentation confirms fifth-gen Tensor Cores with FP4/FP8/FP6 low-precision support for deep learning workloads, and community discussion references RTX 5090 use for local LLM inference (though cost concerns are raised). However, there is no evidence of explicit support/documentation for named serving stacks like TensorRT-LLM, vLLM, or llama.cpp on this specific part. Missing for 10: documented compatibility with TensorRT-LLM, vLLM, or llama.cpp serving frameworks, benchmark data showing actual inference throughput on this GPU.
- [claimed-docs] “Fifth-generation Tensor Cores deliver up to 3x higher throughput for deep learning, with new FP4 support for massive performance gains along…”
- [community] “"If the prices for the RTX 5090 remain at 3500€, they will likely remain insignificant for the DIY crowd" for local LLM use compared to used…”
- [claimed-docs] “The GeForce RTX 5090 is powered by the NVIDIA Blackwell architecture and equipped with 32 GB of super-fast GDDR7 memory, so you can do it al…”
ROCm docs explicitly document serving-stack support for this Radeon line (vLLM 'Full support', llama.cpp 'Supported for efficient inference', plus PyTorch/TensorFlow/JAX/ONNX), but there is no mention of low-precision FP8/FP4 inference formats for the RX 9070 XT, and TensorRT-LLM is an NVIDIA-only stack so is not applicable here. Missing for 10: explicit FP8/FP4 quantization support documentation for this card, independent hands-on benchmarks confirming these serving stacks actually run well on RX 9070 XT.
- [claimed-docs] “**vLLM**: Full support.”
- [claimed-docs] “**Llama.cpp**: Supported for efficient inference.”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Linux® PyTorch, TensorFlow, JAX, ONNX”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Windows® PyTorch”
Tensor specs
ml engineerSize training and inference from published tensor throughput — TFLOPS or TOPS with precision and sparsity stated, not a bare marketing number
weight 3 · round to GeForce RTX 5090Docs mention fifth-gen Tensor Cores with specific precision support (FP4, TF32, BF16, FP16, FP8, FP6) and a relative claim of 'up to 3x higher throughput,' but no evidence gives an absolute TFLOPS/TOPS figure paired with sparsity state — the story explicitly wants a non-bare number with precision AND sparsity documented. Missing for 10: dense vs sparse TOPS/TFLOPS tables per precision, official spec-sheet numeric throughput figures, and independent benchmark corroboration of those numbers for ML workload sizing.
- [claimed-docs] “Fifth-generation Tensor Cores deliver up to 3x higher throughput for deep learning, with new FP4 support for massive performance gains along…”
- [probe] “PROBE runtime (recorded 2026-09-15): NVIDIA's RTX 5090 product/spec page answered a keyless curl and names the part — the vendor spec surfac…”
Radeon RX 9070 XTnone0/10The evidence pack covers ROCm software support, framework compatibility, and general features, but contains no published TFLOPS/TOPS figures for the RX 9070 XT with precision (FP16/FP32/INT8) or sparsity stated. Missing for 10: any tensor throughput numbers, precision breakdown, sparsity conditions, or comparison to a baseline needed for training/inference sizing.
- [claimed-docs] “OS Support Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit”
- [claimed-docs] “AV1 Decode Yes AV1 Encode Yes”
Automation depth — how much of the product can run unattendedAutomation depth
How much of the product can run unattended
ai-native userPerform bulk operations across many items at once
weight 2 · round drawnGeForce RTX 5090none0/10The 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.)
ai-native userSchedule recurring jobs or workflows
weight 2 · round drawnGeForce RTX 5090none0/10The 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.)
Creator media — stories about creator media in this arenaCreator media
Stories about creator media in this arena
Media engines
creatorHardware media engines and creator-app acceleration are documented — AV1/HEVC encoders, and professional or ISV-certified driver support where the vendor claims it
weight 2 · round to Radeon RX 9070 XTGeForce RTX 5090none0/10The evidence pack covers CUDA, DLSS, RT Cores, and gaming features but contains no documentation of NVENC/AV1/HEVC hardware encoders, Studio Driver certification, or ISV-certified professional app support for the RTX 5090. Missing for 10: any mention of hardware encoder specs, Studio Driver program, or ISV certification for creator applications.
AMD's product page explicitly documents AV1 encode/decode support for the RX 9070 XT, which is relevant to creator workflows, but there is no mention of HEVC encode/decode support, no ISV-certified professional driver program (that's typically reserved for Radeon Pro cards), and no explicit creator-app acceleration claims (e.g., Premiere, DaVinci Resolve, OBS integration). Missing for 10: HEVC encoder/decoder documentation, ISV/professional certification claims, named creator-app acceleration partnerships or benchmarks.
- [claimed-docs] “AV1 Decode Yes AV1 Encode Yes”
Datacenter scale — stories about datacenter scale in this arenaDatacenter scale
Stories about datacenter scale in this arena
Scale out
ml engineerTrain and serve at datacenter scale on this part — documented high-bandwidth interconnect (NVLink, Infinity Fabric), multi-GPU systems, and rack-scale deployment
weight 3 · round drawnGeForce RTX 5090none0/10The evidence pack contains no mention of NVLink, Infinity Fabric, multi-GPU interconnect, or rack-scale deployment for the RTX 5090; all specs describe a single consumer GPU (GDDR7, PCIe 5.0) and community commentary discusses it only as a DIY/local-LLM card, not datacenter-scale training/serving infrastructure.
- [claimed-docs] “The GeForce RTX 5090 is powered by the NVIDIA Blackwell architecture and equipped with 32 GB of super-fast GDDR7 memory, so you can do it al…”
- [community] “Pros: Fastest GPU around (usually), 32GB GDDR7 on 512-bit bus, PCIe 5.0, potent AI performance. Cons: Driver issues in some games/apps, extr…”
- [community] “"If the prices for the RTX 5090 remain at 3500€, they will likely remain insignificant for the DIY crowd" for local LLM use compared to used…”
Radeon RX 9070 XTnone0/10The evidence only shows ROCm software compatibility and a vague note that the same stack 'also supports' Instinct/CDNA datacenter accelerators, but nothing documents NVLink/Infinity Fabric interconnect, multi-GPU scaling, or rack-scale deployment for the RX 9070 XT itself, which is a single consumer desktop card without such interconnects. missing for 10: any documentation of multi-GPU interconnect (NVLink/Infinity Fabric) for this card, multi-GPU system support, rack-scale deployment guidance.
- [claimed-docs] “The same ROCm™ stack that powers your desktop development on RDNA™ architecture GPUs also supports AMD Instinct™ accelerators on CDNA™ archi…”
- [claimed-docs] “This unified platform creates a seamless migration path, allowing you to develop applications locally and deploy them at scale with confiden…”
Driver openness — stories about driver openness in this arenaDriver openness
Stories about driver openness in this arena
Linux support
developerLinux is a first-class citizen for this GPU — documented Linux driver releases and independent Linux testing of this part
weight 2 · round to Radeon RX 9070 XTIndependent Phoronix benchmarking confirms the RTX 5090 is tested and functional on Linux across 60+ compute workloads, showing real-world Linux usability, but the evidence pack contains no first-party documentation of dedicated Linux driver releases, changelogs, or Linux-specific support pages from NVIDIA. Missing for 10: documented NVIDIA Linux driver release notes/changelog, official Linux support/compatibility docs, broader independent Linux gaming/compute test corroboration beyond one Phoronix reference.
- [community] “Phoronix Linux benchmarks found the GeForce RTX 5090 delivering 1.42x the performance of the RTX 4090 across 60+ compute benchmarks, but on …”
AMD's official docs explicitly list Linux x86-64 as a supported OS and provide detailed ROCm Linux documentation for PyTorch/TensorFlow/JAX/ONNX, vLLM, and llama.cpp on Radeon 9000-series GPUs, showing genuine first-class Linux driver/software support. However, the pack lacks independent hands-on Linux testing or benchmarks of the RX 9070 XT specifically (the only community citation is a general Windows-oriented performance review, not Linux-focused). Missing for 10: independent/third-party Linux driver stability or performance testing of this specific card, and any community confirmation of ROCm functionality on this GPU outside vendor docs.
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Linux® PyTorch, TensorFlow, JAX, ONNX”
- [claimed-docs] “OS Support Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit”
- [claimed-docs] “**vLLM**: Full support.”
- [claimed-docs] “**Llama.cpp**: Supported for efficient inference.”
- [claimed-docs] “As a primarily open-source ecosystem, ROCm™ gives you the freedom to inspect, customize, and tailor the software stack to your specific need…”
Open drivers
developerRun this GPU on an open driver — open-source kernel modules or upstream Linux support documented by the vendor
weight 2 · round to Radeon RX 9070 XTGeForce RTX 5090none0/10The evidence pack contains no mention of open-source kernel modules (e.g., NVIDIA's open GPU kernel module project) or upstream Linux driver support for the RTX 5090; all driver-related community mentions are about closed-driver bugs/issues, not open-source availability. This is a fair axis for a GPU (vendors like NVIDIA do ship open kernel modules), but nothing in the evidence documents it for this card.
AMD documents ROCm as a primarily open-source stack with Linux support and lists RX 9000-series compatibility, and Linux x86 64-bit OS support is confirmed, but there is no explicit vendor documentation of open-source kernel driver components (e.g., amdgpu upstream kernel module) specific to RX 9070 XT or a clear statement of which parts of the stack are closed-source firmware/blobs. missing for 10: explicit vendor confirmation of upstream open-source kernel module support for this specific GPU, independent/hands-on corroboration of open driver functioning on mainline Linux kernels.
- [claimed-docs] “As a primarily open-source ecosystem, ROCm™ gives you the freedom to inspect, customize, and tailor the software stack to your specific need…”
- [claimed-docs] “OS Support Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Linux® PyTorch, TensorFlow, JAX, ONNX”
Gaming performance — stories about gaming performance in this arenaGaming performance
Stories about gaming performance in this arena
4k gaming
gamerThis card drives high-refresh 4K gaming — vendor performance claims corroborated by independent game benchmarks
weight 3 · round to GeForce RTX 5090Independent Tom's Hardware and Phoronix benchmarks corroborate the RTX 5090 as the fastest consumer GPU (1.42x over 4090), supporting vendor claims of top-tier gaming performance, and NVIDIA's DLSS/Reflex/path-tracing feature docs target high-refresh 4K use cases. However, community evidence raises real caveats: reviewers flag Founders Edition thermal issues, similar-or-worse perf-per-watt vs 4080/4090, and specifically critique Multi Frame Generation as 'marketing' since input sampling doesn't scale with the reported FPS boost, undercutting the headline frame-rate claims. missing for 10: dedicated 4K high-refresh benchmark suites (e.g. specific FPS-at-4K numbers across many AAA titles), resolution of the MFG input-latency criticism, and confirmation thermal/efficiency issues don't limit sustained high-refresh performance.
- [claimed-docs] “Powered by GeForce RTX 50 Series and fifth-generation Tensor Cores, new DLSS Multi Frame Generation boosts FPS by using AI to generate up to…”
- [claimed-docs] “Reflex technologies optimize the graphics pipeline for ultimate responsiveness, providing faster target acquisition, quicker reaction times,…”
- [claimed-docs] “The NVIDIA Blackwell architecture unlocks the game-changing realism of path tracing. Experience cinematic quality visuals at unprecedented s…”
- [community] “Pros: Fastest GPU around (usually), 32GB GDDR7 on 512-bit bus, PCIe 5.0, potent AI performance. Cons: Driver issues in some games/apps, extr…”
- [community] “We've now docked half a star, due to concerns specifically with the Founders Edition running hot.”
- [community] “MFG as an example running at 240 FPS would mean user input only gets sampled at 60 FPS. That's not the same as a game running at 240 FPS nat…”
- [community] “Phoronix Linux benchmarks found the GeForce RTX 5090 delivering 1.42x the performance of the RTX 4090 across 60+ compute benchmarks, but on …”
Only one independent source (TechPowerUp) confirms the RX 9070 XT delivers competitive raster/ray-tracing performance versus similarly priced NVIDIA cards, but it doesn't specifically address 4K high-refresh benchmarks or vendor FPS claims. Missing for 10: explicit vendor 4K/144Hz+ performance claims, detailed independent 4K benchmark numbers across multiple games, and confirmation of sustained high-refresh framerates.
- [community] “The RX 9070 XT offers competitive performance with similarly priced NVIDIA options in both raster and ray tracing, at a starting price of $6…”
Upscaling
gamerAI upscaling and frame generation are supported on this card — the DLSS or FSR generation is documented for this part, with broad game support
weight 2 · round to GeForce RTX 5090NVIDIA's official docs detail DLSS 4 with Multi Frame Generation, Super Resolution, Ray Reconstruction, and DLAA on RTX 5090/50-series, plus broad game rollout via the NVIDIA app supporting hundreds of titles. Independent review (Tom's Hardware) confirms these features work in practice, though it flags marketing caveats around Multi Frame Generation's input latency implications. Missing for 10: independent per-game compatibility list/count and more third-party benchmarking corroboration beyond one review.
- [claimed-docs] “Powered by GeForce RTX 50 Series and fifth-generation Tensor Cores, new DLSS Multi Frame Generation boosts FPS by using AI to generate up to…”
- [claimed-docs] “Dynamically adjust your multiplier to maximize smoothness across different games and scenes on GeForce RTX 50 Series GPUs.”
- [claimed-docs] “Enhances image quality by using AI to generate additional pixels for intensive ray-traced scenes.”
- [claimed-docs] “Boosts performance by using AI to output higher-resolution frames from a lower-resolution input.”
- [claimed-docs] “Provides higher image quality with an AI-based anti-aliasing technique. DLAA uses the same Super Resolution technology developed for DLSS, c…”
- [claimed-docs] “With the NVIDIA app you can update hundreds of games to use the latest DLSS features including Multi Frame Generation, and the newest AI mod…”
- [community] “Pros: Fastest GPU around (usually), 32GB GDDR7 on 512-bit bus, PCIe 5.0, potent AI performance. Cons: Driver issues in some games/apps, extr…”
- [community] “MFG as an example running at 240 FPS would mean user input only gets sampled at 60 FPS. That's not the same as a game running at 240 FPS nat…”
AMD's product page confirms FSR ("Redstone") support on the RX 9070 XT, and third-party review corroborates strong raster/ray-tracing performance, but the evidence pack lacks detail on frame-generation specifics or a documented list of supported games. missing for 10: explicit frame-generation (FSR 3/4) feature naming, game compatibility list/count, independent hands-on upscaling benchmarks.
- [claimed-docs] “AMD FSR™ "Redstone"”
- [community] “The RX 9070 XT offers competitive performance with similarly priced NVIDIA options in both raster and ray tracing, at a starting price of $6…”
Memory vram — stories about memory vram in this arenaMemory vram
Stories about memory vram in this arena
Llm memory
ai-native userRun a 70B-class quantized LLM on this GPU — published VRAM capacity and memory bandwidth that make local or single-node inference practical
weight 3 · round to GeForce RTX 5090NVIDIA publishes the RTX 5090's 32GB GDDR7 VRAM (docs-18) and strong Tensor Core throughput (docs-22), which is enough for some 70B-class models at aggressive quantization, but the evidence never states a specific memory-bandwidth figure and community discussion (comm-5) notes the GPU's price makes it 'insignificant for the DIY crowd' compared to used 3090s or Mac unified memory for local LLM work, undercutting the 'practical' framing. Missing for 10: published memory-bandwidth spec (GB/s), explicit vendor guidance on running 70B-class quantized models, and independent benchmarks confirming inference throughput at that scale.
- [claimed-docs] “The GeForce RTX 5090 is powered by the NVIDIA Blackwell architecture and equipped with 32 GB of super-fast GDDR7 memory, so you can do it al…”
- [claimed-docs] “Fifth-generation Tensor Cores deliver up to 3x higher throughput for deep learning, with new FP4 support for massive performance gains along…”
- [community] “"If the prices for the RTX 5090 remain at 3500€, they will likely remain insignificant for the DIY crowd" for local LLM use compared to used…”
Radeon RX 9070 XTnone0/10The evidence pack never states the RX 9070 XT's actual VRAM capacity or memory bandwidth; the one VRAM figure mentioned ("up to 48GB") is a generic Radeon workstation-GPU claim, not this card's spec, and there is no claim or benchmark showing a 70B-class quantized model running on this GPU. Software support (ROCm, vLLM, llama.cpp) is documented but doesn't substitute for the missing capacity/bandwidth evidence needed to judge practicality for 70B inference.
- [claimed-docs] “**vLLM**: Full support.”
- [claimed-docs] “**Llama.cpp**: Supported for efficient inference.”
- [claimed-docs] “A local workstation equipped with a Radeon™ GPU, featuring up to 48GB of VRAM, offers a secure and economical alternative to relying solely …”
Memory spec
ml engineerMemory specs are published in full for this exact part — capacity, memory type, bus width, and bandwidth
weight 2 · round to GeForce RTX 5090NVIDIA's own product page confirms capacity (32GB) and memory type (GDDR7), and community review corroborates a 512-bit bus, but no evidence pack item states the exact memory bandwidth figure (GB/s) for this part. Missing for 10: an explicit published bandwidth number (GB/s) from a first-party spec sheet.
- [claimed-docs] “The GeForce RTX 5090 is powered by the NVIDIA Blackwell architecture and equipped with 32 GB of super-fast GDDR7 memory, so you can do it al…”
- [community] “Pros: Fastest GPU around (usually), 32GB GDDR7 on 512-bit bus, PCIe 5.0, potent AI performance. Cons: Driver issues in some games/apps, extr…”
- [probe] “PROBE runtime (recorded 2026-09-15): NVIDIA's RTX 5090 product/spec page answered a keyless curl and names the part — the vendor spec surfac…”
Openness — open source, data portability, and self-hosting storiesOpenness
Open source, data portability, and self-hosting stories
ai-native userRead the product's source under an open license
weight 2 · round to Radeon RX 9070 XTGeForce RTX 5090none0/10The 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.)
AMD's ROCm docs claim the stack is a 'primarily open-source ecosystem' giving users freedom to inspect and customize the software, and this is corroborated by an accessible llms.txt docs endpoint, but this is the accompanying software stack, not the GPU's actual hardware/firmware source, and 'primarily' implies some closed-source components remain undisclosed. Missing for 10: explicit repository/license pointer for the actual open-sourced source code, confirmation of what portions (drivers, firmware) are closed, and independent hands-on verification of source availability.
- [claimed-docs] “As a primarily open-source ecosystem, ROCm™ gives you the freedom to inspect, customize, and tailor the software stack to your specific need…”
- [probe] “PROBE llms.txt: HTTP 200 at https://rocm.docs.amd.com/llms.txt # ROCm documentation > Note: ROCm documentation is split across multiple pro…”
ai-native userSelf-host the core product
weight 3 · round to Radeon RX 9070 XTGeForce RTX 5090none0/10The 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.)
As a discrete GPU, the RX 9070 XT is inherently self-hosted hardware; AMD's docs explicitly position a local Radeon workstation as a secure, economical alternative to cloud-based AI solutions, backed by open ROCm stack support for PyTorch/TensorFlow/JAX/ONNX/vLLM/llama.cpp on both Windows and Linux. Missing for 10: independent hands-on validation of a fully self-hosted AI stack setup, and more detailed first-party self-hosting/deployment guides beyond general ROCm docs.
- [claimed-docs] “A local workstation equipped with a Radeon™ GPU, featuring up to 48GB of VRAM, offers a secure and economical alternative to relying solely …”
- [claimed-docs] “This unified platform creates a seamless migration path, allowing you to develop applications locally and deploy them at scale with confiden…”
- [claimed-docs] “The same ROCm™ stack that powers your desktop development on RDNA™ architecture GPUs also supports AMD Instinct™ accelerators on CDNA™ archi…”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Linux® PyTorch, TensorFlow, JAX, ONNX”
- [claimed-docs] “**vLLM**: Full support.”
- [claimed-docs] “**Llama.cpp**: Supported for efficient inference.”
Power cooling — stories about power cooling in this arenaPower cooling
Stories about power cooling in this arena
Efficiency
ml engineerSustained workloads are power-efficient on this part — documented power envelopes with independent performance-per-watt testing
weight 2 · round to GeForce RTX 5090GeForce RTX 5090disputedcontradicted5/10NVIDIA's spec page exposes board power/TOPS figures (probe-rt-1) but no explicit efficiency/perf-per-watt marketing claim is cited; independent testing (Phoronix via HN, comm-6/7/8) directly contradicts any efficiency narrative, finding RTX 5090 perf-per-watt is similar to or worse than the 4080/4090 despite higher absolute performance, and Tom's Hardware also flags thermal/power concerns (comm-2). Missing for 10: a documented power-envelope/TDP spec sheet from NVIDIA explicitly framed for ML workloads, and independent perf-per-watt benchmarks that confirm rather than undercut efficiency gains.
- [probe] “PROBE runtime (recorded 2026-09-15): NVIDIA's RTX 5090 product/spec page answered a keyless curl and names the part — the vendor spec surfac…”
- [community] “TL;DR; performance isn't bad, but perf per Watt isn't better than 4080 or 4090 and can even be significantly lower than 4090 in certain cont…”
- [community] “"the 4080 Super doing well compared to the 5080 and 5090... seems to have better performance per watt ratio than them while also having some…”
- [community] “Phoronix Linux benchmarks found the GeForce RTX 5090 delivering 1.42x the performance of the RTX 4090 across 60+ compute benchmarks, but on …”
- [community] “We've now docked half a star, due to concerns specifically with the Founders Edition running hot.”
Radeon RX 9070 XTnone0/10Evidence covers ROCm software support, framework compatibility, and general GPU features, but there is no documented power envelope data or independent performance-per-watt testing for sustained ML workloads on this card. missing for 10: TDP/power envelope specs for sustained ML loads, independent perf-per-watt benchmarks, thermal/power throttling behavior under long-running compute jobs.
Psu planning
gamerSpec a build around published board power — TDP/TGP, connector requirements, and cooling guidance for this exact card
weight 2 · round to GeForce RTX 5090GeForce RTX 5090disputedcontradicted4/10Evidence confirms NVIDIA's official 5090 page exposes board-power and spec data (probe-rt-1) and the community references the 12VHPWR/12V-2x6 connector spec, but no evidence pack item actually cites the published TDP/TGP wattage, PSU wattage recommendation, or explicit cooling/case guidance for this exact card. More importantly, community evidence (comm-4) documents real-world 12VHPWR connector melting incidents tied to the card's power-connector design, and comm-2 reports the Founders Edition running hot — concretely undermining confidence that a build spec'd purely around the vendor's connector/cooling guidance will be reliable. missing for 10: explicit first-party TDP/TGP wattage figure, official PSU/connector spec sheet, first-party cooling/case airflow guidance, resolution of the connector-melting safety concern.
- [probe] “PROBE runtime (recorded 2026-09-15): NVIDIA's RTX 5090 product/spec page answered a keyless curl and names the part — the vendor spec surfac…”
- [community] “Discussion centers on a reported RTX 5090 FE 12VHPWR connector melting; users debate that the FE's single 12V bus-bar design relies purely o…”
- [community] “We've now docked half a star, due to concerns specifically with the Founders Edition running hot.”
- [claimed-docs] “The GeForce RTX 5090 is powered by the NVIDIA Blackwell architecture and equipped with 32 GB of super-fast GDDR7 memory, so you can do it al…”
Radeon RX 9070 XTnone0/10The evidence pack covers ROCm/software ecosystem support, AMD Software features, and general performance comparison, but contains no mention of TDP/TGP figures, PCIe power connector requirements, or cooling/case guidance for the RX 9070 XT. This is a fair axis for a discrete GPU, but nothing in the pack substantiates it.
Privacy posture — data-handling and privacy storiesPrivacy posture
Data-handling and privacy stories
ai-native userPrevent my data from being used to train AI models
weight 3 · round drawnGeForce RTX 5090none0/10The 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.)
Software toolchain — stories about software toolchain in this arenaSoftware toolchain
Stories about software toolchain in this arena
Compute stack
developerShip GPU-compute workloads on the vendor's toolchain — CUDA or ROCm/HIP documentation lists this part as a supported target
weight 3 · round drawnNVIDIA's CUDA Toolkit docs explicitly target GeForce RTX GPUs including Blackwell-architecture cards like the 5090, and independent Phoronix benchmarks (cited via comm-8) confirm real compute workloads running on the 5090 with 1.42x uplift over the 4090 across 60+ compute benchmarks, corroborating that it functions as a genuine CUDA compute target. Missing for 10: no explicit CUDA compute-capability/SM version listing naming '5090' directly in vendor docs, and no independent report specifically validating professional GPU-compute (non-gaming) toolchains beyond Phoronix's Linux compute suite.
- [claimed-docs] “The NVIDIA® CUDA® Toolkit provides a development environment for creating high-performance, GPU-accelerated applications.”
- [claimed-docs] “CUDA Tile C++ is an expression of the CUDA Tile programming model in C++. It's built on top of the CUDA Tile IR specification and allows you…”
- [claimed-docs] “cuTile Python is an expression of the CUDA Tile programming model in Python.”
- [claimed-docs] “Fifth-generation Tensor Cores deliver up to 3x higher throughput for deep learning, with new FP4 support for massive performance gains along…”
- [community] “Phoronix Linux benchmarks found the GeForce RTX 5090 delivering 1.42x the performance of the RTX 4090 across 60+ compute benchmarks, but on …”
ROCm documentation explicitly lists RX 9000 series (including 9070 XT) as a supported target on both Windows and Linux, with PyTorch, TensorFlow, JAX, ONNX, vLLM, and Llama.cpp support, plus a stated migration path to AMD Instinct datacenter GPUs. missing for 10: independent hands-on developer corroboration of ROCm workflows on this specific card beyond vendor docs, and no mention of CUDA compatibility layer maturity/limitations.
- [claimed-docs] “PyTorch on Windows updated with ROCm 7.2.1 on AMD Radeon graphics products and AMD Ryzen AI processors.”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Linux® PyTorch, TensorFlow, JAX, ONNX”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Windows® PyTorch”
- [claimed-docs] “**vLLM**: Full support.”
- [claimed-docs] “**Llama.cpp**: Supported for efficient inference.”
- [claimed-docs] “The same ROCm™ stack that powers your desktop development on RDNA™ architecture GPUs also supports AMD Instinct™ accelerators on CDNA™ archi…”
Frameworks
ml engineerPyTorch and mainstream ML frameworks run on this GPU through officially documented builds and support matrices
weight 2 · round to Radeon RX 9070 XTEvidence confirms NVIDIA ships a general CUDA Toolkit and Tensor Core architecture details relevant to deep learning, but there is no explicit official PyTorch/TensorFlow support matrix, compute-capability listing, or documented install path referencing RTX 5090/Blackwell specifically; a community post even flags 5090 as impractical for local LLM work due to price, not toolchain issues. missing for 10: explicit PyTorch build/support-matrix docs naming RTX 5090 or Blackwell (sm_120) compute capability, cuDNN/cuBLAS version compatibility notes, and independent confirmation of PyTorch working out-of-the-box on the card.
- [claimed-docs] “The NVIDIA® CUDA® Toolkit provides a development environment for creating high-performance, GPU-accelerated applications.”
- [claimed-docs] “Fifth-generation Tensor Cores deliver up to 3x higher throughput for deep learning, with new FP4 support for massive performance gains along…”
- [community] “"If the prices for the RTX 5090 remain at 3500€, they will likely remain insignificant for the DIY crowd" for local LLM use compared to used…”
AMD's official ROCm docs explicitly list RX 9000 series support for PyTorch (Windows and Linux), TensorFlow, JAX, ONNX, vLLM, and Llama.cpp, with a documented support matrix and OS support (Windows/Linux) confirmed on AMD's product page. Missing for 10: independent hands-on confirmation of install success/version compatibility and no community corroboration of real-world PyTorch training/inference workflows on this specific card.
- [claimed-docs] “PyTorch on Windows updated with ROCm 7.2.1 on AMD Radeon graphics products and AMD Ryzen AI processors.”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Linux® PyTorch, TensorFlow, JAX, ONNX”
- [claimed-docs] “Radeon™ GPUs (9000 & select 7000 Series) Windows® PyTorch”
- [claimed-docs] “**vLLM**: Full support.”
- [claimed-docs] “**Llama.cpp**: Supported for efficient inference.”
- [claimed-docs] “OS Support Windows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit”
Not comparable on these axes
ai-native userPlug MCP servers into this product so it can use their tools
weight 3 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a GPU hardware product, not an agent or platform that consumes MCP servers; plugging in MCP tool servers is a software/agent-layer capability entirely outside this product's category.
ai-native userConnect an agent via an official MCP server
weight 3 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer GPU hardware product, not an agentic software platform or service; connecting agents via an official MCP server is outside its product category — a category error, not a missing feature.
ai-native userIssue scoped/least-privilege API credentials for an agent
weight 2 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer GPU hardware product, not an API/service platform; scoped API credential issuance is a category error for this product type.
ai-native userSubscribe to events via webhooks
weight 2 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer GPU hardware product, not a service or platform with event-driven subscription APIs; webhooks are a category error for this kind of product.
ai-native userSet up automations that run autonomously in the background
weight 2 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer GPU hardware product, not an automation/agent platform; setting up background-running automations is an application/software-layer capability entirely outside the scope of a graphics card, making this a category error rather than a missing feature.
Radeon RX 9070 XTn/aThis is a GPU hardware product; autonomous background automation is an application/software-orchestration capability, not something a graphics card ships itself. The evidence only covers driver/software stack support (ROCm, PyTorch) for running AI workloads, not automation/agent orchestration features.
ai-native userExplore an interactive API reference with runnable examples
weight 2 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer GPU hardware product, not an API/SDK service; an interactive API reference with runnable examples is a category error for this axis. Evidence shows no such interactive reference exists (openapi/docs probes return 404), reinforcing this is not applicable to the hardware product itself.
Radeon RX 9070 XTn/aThe RX 9070 XT is a physical GPU product, not an API/SaaS product with a callable API reference; this axis targets developer-facing API docs and does not apply to a hardware product's own interface (ROCm software docs are a separate ecosystem artifact, not an interactive API reference for the GPU itself).
ai-native userDownload a machine-readable API spec (OpenAPI or equivalent)
weight 2 · not comparableGeForce RTX 5090none0/10This is a consumer GPU product; while NVIDIA's developer portal exists, an explicit probe found no OpenAPI/machine-readable API spec (all candidate paths 404), so there's no evidence of a downloadable machine-readable API spec.
- [probe] “PROBE openapi: all candidate paths 404 (https://developer.nvidia.com/openapi.json, https://developer.nvidia.com/swagger.json, https://develo…”
ai-native userTest against a sandbox environment without touching production data
weight 1 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer GPU hardware product, not a software/service platform with sandbox vs production environments; testing against sandbox data is a category error for this product type.
ai-native userRely on versioned APIs with a documented deprecation policy
weight 2 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer GPU hardware product, not an API/service platform; versioned APIs and deprecation policies are not applicable to this product category.
ai-native userDefine rules that trigger actions automatically on events
weight 3 · not comparableGeForce RTX 5090n/aThe GeForce RTX 5090 is a consumer GPU hardware product, not an automation/workflow platform; defining event-triggered rules is a category error for this axis - no evidence pack material addresses rule-based automation.
ai-native userVersion, review, and roll back my automations
weight 1 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer GPU hardware product, not an automation/workflow platform; versioning, reviewing, and rolling back automations is a software/orchestration concept entirely outside a GPU's product category.
ai-native userDo everything through the API that I can do in the UI
weight 2 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer graphics hardware product, not a UI/API software product; there is no 'UI' vs 'API' parity concept applicable to a physical GPU—this axis is a category error for this product type.
ai-native userExport all of my data in open formats and leave
weight 3 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a GPU hardware product with no user account, data storage, or data-export concept; 'exporting data in open formats and leaving' is a SaaS/platform lock-in axis that doesn't apply to a physical graphics card.
ai-native userChoose where my data is stored (region/residency)
weight 2 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer GPU hardware product with no data-hosting or cloud service component; data residency/region storage choice is not an applicable axis for local hardware.
ai-native userControl data retention and deletion
weight 2 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer GPU hardware product, not a data-processing service or platform that collects/retains user data; data retention/deletion controls are not a meaningful axis for a piece of silicon.
ai-native userOpt out of telemetry and usage tracking
weight 2 · not comparableGeForce RTX 5090n/aThe RTX 5090 is a consumer graphics card, not a service or software platform that collects user telemetry to opt out of; this privacy-posture axis about opting out of usage tracking applies to software/SaaS products, not GPU hardware.