In this article
194 model profiles and grouped variants, plus six guides: NVIDIA, Radeon and Instinct memory, software boundaries, cooling and workload qualification in English and Russian.
How to read the catalogue
This index covers desktop, professional and server GPU families relevant to evaluating LLM inference or rendering. Each profile distinguishes published specifications from workload qualification and links to primary documentation. A model can be a rendering candidate, an inference candidate, a conditional legacy experiment or a published future platform; inclusion does not mean all four. Capacity is not a model-size guarantee. The entries are model profiles, sometimes grouping official memory or cooling variants, rather than a claim that every partner board, laptop configuration and historical accelerator has been inventoried. Specifications were reviewed on 2 November 2026.
NVIDIA GeForce and TITAN
GeForce and TITAN profiles separate reference memory variants and architecture generations. A consumer GPU can be useful for a qualified workstation workload without becoming a validated server installation. Modern vLLM requirements and renderer eligibility differ, so the legacy entries do not inherit the capabilities of a new RTX card. Regional D variants and officially documented memory configurations retain their own identities. The table links to the detailed assessment; its capacity column must be read with the scope notes in that article.
NVIDIA RTX PRO, RTX A and Quadro
Professional names need their full suffix: Workstation, Max-Q, Server, SFF and passive variants can change mechanics and cooling while sharing a memory capacity. The group also includes low-memory families where a combined profile is more informative than near-identical short articles. An official compute-capability entry is not a renderer or inference-engine certification. Read the individual profile for CUDA generation, driver constraints, display assumptions and unresolved fields before treating a workstation board as an unattended server resource.
AMD Radeon RX, AI PRO and Radeon PRO
Radeon profiles distinguish exact ROCm SKU listings from the RDNA eligibility boundary for Cycles HIP. Those are separate questions. The catalogue also records vendor datacenter-validation limitations for workstation families and treats Vega-era devices as legacy experiments where appropriate. RX 9070 GRE has conflicting capacity information across checked AMD documents; its profile uses the dedicated product-page value and discloses the discrepancy. A V340 board total is explicitly not presented as one unified device allocation.
Datacenter, module and legacy accelerator profiles
This group includes graphics-capable server boards, compute accelerators, multi-GPU assemblies and platform modules. Memory scope is especially important here: per-GPU, per-GCD, per-board, shared APU and node totals cannot be substituted for one another. Instinct CDNA compute support does not imply current Cycles HIP eligibility. Rubin and MI455X entries describe published platform specifications without asserting availability. MI430X is explicitly a 2027 roadmap entry. An archived OEM product marked withdrawn is not represented as a current purchasing offer.
| Model / group | Published memory scope | Segment |
|---|---|---|
| NVIDIA L40S | 48 GB | server |
| NVIDIA L40 | 48 GB | server |
| NVIDIA L4 | 24 GB | server |
| NVIDIA L20 | 48 GB | server |
| NVIDIA A40 | 48 GB | server |
| NVIDIA A10 | 24 GB | server |
| NVIDIA A16 | 4 × 16 GB; 64 GB per board | server |
| NVIDIA A2 | 16 GB | server |
| NVIDIA A30 | 24 GB | server |
| NVIDIA A100 | 40/80 GB | server |
| NVIDIA A800 | 80 GB | server |
| NVIDIA H100 | 80/94 GB | server |
| NVIDIA H200 | 141 GB | server |
| NVIDIA H800 | 80/94 GB | server |
| NVIDIA H20 | 96/141 GB | server |
| NVIDIA B200 | 180 GB usable HGX | server |
| NVIDIA B300 | 288 GB | server |
| NVIDIA Rubin GPU | 288 GB | platform specification |
| NVIDIA Tesla V100 | 16/32 GB | server |
| NVIDIA Tesla V100S | 32 GB | server |
| NVIDIA Tesla P100 | 12/16 GB | server |
| NVIDIA Tesla P40 | 24 GB | server |
| NVIDIA Tesla P4 | 8 GB | server |
| NVIDIA Tesla T4 | 16 GB | server |
| NVIDIA Tesla M40 | 12/24 GB | server |
| NVIDIA Tesla M60 | 2 × 8 GB; 16 GB per board | server |
| NVIDIA Tesla M10 | 4 × 8 GB; 32 GB per board | server |
| NVIDIA Tesla M6 and P6 | 8/16 GB | server |
| NVIDIA GeForce GTX 980 Ti | 6 GB | desktop |
| NVIDIA GeForce GTX TITAN X Maxwell | 12 GB | desktop |
| AMD Instinct MI100 | 32 GB | server |
| AMD Instinct MI210 | 64 GB | server |
| AMD Instinct MI250 and MI250X | 2 × 64 GB per OAM | server |
| AMD Instinct MI300X | 192 GB | server |
| AMD Instinct MI300A | 128 GB shared CPU/GPU | server |
| AMD Instinct MI325X | 256 GB | server |
| AMD Instinct MI350P | 144 GB | server |
| AMD Instinct MI350X and MI355X | 288 GB | server |
| AMD Instinct MI455X | 432 GB | platform specification |
| AMD Instinct MI430X | 432 GB | planned 2027 |
| AMD Instinct MI50 and MI60 | 16/32 GB | server |
Scope boundaries and unresolved products
Laptop GPUs, Apple MPX modules, integrated Radeon and broader CPU/GPU systems require separate host-level memory and thermal records. A desktop article must not be reused as their specification. Mining-only boards, unofficial memory modifications and cosmetic partner editions are not separate verified profiles. Full primary board specifications were not established for NVIDIA L2, RTX PRO 6000 D and AMD Instinct MI308X; their capacity is not invented. Older pre-RDNA Radeon and pre-current-engine NVIDIA families are explained in the legacy guide rather than advertised as modern supported production targets. These boundaries make the catalogue auditable instead of claiming universal completeness.
Primary references and interpretation
The scope draws on the NVIDIA compute inventory, AMD graphics specifications and AMD accelerator specifications. Software interpretation uses the ROCm matrix, vLLM requirements and Cycles manual. Individual articles supply the matching product and OEM references. Editorial query phrases are not measured search volumes, and this catalogue does not claim benchmark results for every GPU. Retrivium is developing its compute-provider integration with OpenRouter; these educational profiles are not a competing router catalogue or a declaration of installed inventory.
Related articles: GPU compatibility for LLM inference and Cycles rendering, GPU VRAM, board totals and shared memory: how to read specifications, Desktop GPU, workstation or server accelerator: qualify the host, Legacy GPUs for LLMs and rendering: experimental versus supported, Sharing GPUs between LLM inference and server rendering. GPU catalogue.