https://vram.wiki https://vram.wiki/stats https://vram.wiki/compare https://vram.wiki/badges https://vram.wiki/founders https://vram.wiki/glossary https://vram.wiki/about https://vram.wiki/roadmap https://vram.wiki/patterns https://vram.wiki/papers https://vram.wiki/use-case/teacher-exam-grading-qwen-3090 https://vram.wiki/use-case/lawyer-document-review https://vram.wiki/use-case/coaching-business-dgx-cluster https://vram.wiki/use-case/saas-amd-vllm-harness https://vram.wiki/use-case/ecommerce-store-qwen-27b https://vram.wiki/use-case/strix-halo-homelab-research https://vram.wiki/use-case/rag-ingestion-privacy-questions https://vram.wiki/use-case/desktop-tidy-dashboard-pi https://vram.wiki/use-case/hermes-cross-device-agent-transfer https://vram.wiki/use-case/year-long-life-assistant https://vram.wiki/use-case/dual-dgx-daily-driver-reignited https://vram.wiki/use-case/custom-harness-shipped-kmp-projects https://vram.wiki/use-case/freelancer-multi-rig-nda https://vram.wiki/use-case/hermes-cron-doc-review https://vram.wiki/use-case/mac-m1-max-domain-glossary-coding https://vram.wiki/use-case/batch-tasks-setup-guide https://vram.wiki/use-case/paperless-ocr-rag-pipeline https://vram.wiki/use-case/wedding-planner-email-declutter-agent https://vram.wiki/use-case/m2-ultra-rag-uncensored-hybrid https://vram.wiki/use-case/general-use-trial-not-worth-it https://vram.wiki/use-case/rtx3090-opencode-full-dev https://vram.wiki/use-case/cybersec-binary-analysis https://vram.wiki/use-case/stock-market-research-gemma https://vram.wiki/use-case/macbook-cline-homelab-gpt-review https://vram.wiki/use-case/dual-3060-pr-review-support https://vram.wiki/use-case/company-automation-gemma-r9700 https://vram.wiki/use-case/coding-rig-vs-claude-max https://vram.wiki/use-case/router-llama-cpp-opencode https://vram.wiki/use-case/media-tagging-8gb-3070ti https://vram.wiki/use-case/receipt-tracking-nas-pipeline https://vram.wiki/use-case/email-triage-dictation-assistant https://vram.wiki/use-case/medical-report-drafting-privacy https://vram.wiki/use-case/strix-halo-zero-cloud-devops https://vram.wiki/use-case/financial-reports-1500-companies https://vram.wiki/use-case/telegram-multi-agent-personal-ops-strix-halo https://vram.wiki/use-case/four-dgx-spark-vllm-agent-sessions https://vram.wiki/use-case/on-prem-agent-stack-product-team https://vram.wiki/use-case/research-report-autofill-multimodal-rag-4060 https://vram.wiki/use-case/aerospace-itar-document-parser-rag https://vram.wiki/use-case/dynatrace-sre-custom-apps-qwen https://vram.wiki/use-case/mac-m4-pro-document-workflow-app https://vram.wiki/use-case/pentest-code-review-local-m5-strix-halo https://vram.wiki/use-case/local-first-portfolio-tracker-ninfer https://vram.wiki/use-case/construction-estimating-7900xtx-dgx-sparks https://vram.wiki/use-case/qwen-27b-524k-context-godot-game https://vram.wiki/use-case/mcu-esp32-agent-programming-serial-feedback https://vram.wiki/use-case/qwen3-coder-30b-strix-halo-swe https://vram.wiki/use-case/fable-qwen-ticket-implementation https://vram.wiki/use-case/home-assistant-brain-stt https://vram.wiki/use-case/skeptic-real-work-still-claude https://vram.wiki/use-case/deepseek-vision-98-percent-pi-agent https://vram.wiki/use-case/enterprise-a100-b200-sglang https://vram.wiki/use-case/hipaa-dgx-spark-consultant https://vram.wiki/use-case/batch-doc-comment-summarization https://vram.wiki/use-case/rtx5070-hermes-first-productive https://vram.wiki/use-case/qwen36-switch-overthinking-cost https://vram.wiki/use-case/inner-agent-testing-r9700 https://vram.wiki/use-case/subagent-verified-no-file-autonomy https://vram.wiki/use-case/intel-arc-nuc-background-coding https://vram.wiki/use-case/homelab-rag-troubleshooting-advisor https://vram.wiki/use-case/k3s-homelab-ops-hermes https://vram.wiki/use-case/business-wiki-middleware-cancelled-subs https://vram.wiki/use-case/gta-roleplay-document-processing https://vram.wiki/use-case/surveillance-vision-qwen-vl https://vram.wiki/use-case/codex-peer-review-hermes-agent https://vram.wiki/use-case/app-dev-orca-opencode-bionic https://vram.wiki/use-case/call-transcription-rejects-coding https://vram.wiki/use-case/cardiologist-referral-triage https://vram.wiki/use-case/rtx4070-partial-honest-coding https://vram.wiki/use-case/file-organizer-ocr-vision-mac https://vram.wiki/use-case/privacy-prefilter-before-frontier https://vram.wiki/use-case/researcher-m5-dgx-spark-hermes https://vram.wiki/use-case/cpu-only-old-hardware-agent https://vram.wiki/use-case/mobile-phone-daily-driver https://vram.wiki/use-case/igpu-light-tasks-kept-cloud-sub https://vram.wiki/use-case/teacher-robotics-interactive-lessons https://vram.wiki/use-case/skillware-2070-local-automation https://vram.wiki/use-case/ling-opencode-integration-tests https://vram.wiki/use-case/mac-m5-local-scripting https://vram.wiki/use-case/internal-tools-1tb-ecc-server https://vram.wiki/use-case/local-life-finance-planning-assistant https://vram.wiki/use-case/book-editing-dual-5060ti https://vram.wiki/use-case/jetson-mail-inventory-spending-pipeline https://vram.wiki/use-case/8gb-5060-gpu-offload-tuning https://vram.wiki/use-case/r9700-lucebox-optimized-coding https://vram.wiki/use-case/gtx-pro-6000-subagent-dropin https://vram.wiki/use-case/local-only-embeddings-tts-stt-strix-halo https://vram.wiki/use-case/minimum-local-agent-hardware-strix-r9700 https://vram.wiki/use-case/academic-pdf-rag-pipeline-5060ti https://vram.wiki/use-case/pro-dev-prefers-local-keeps-loop https://vram.wiki/use-case/hermes-hobby-inventory-price-watch https://vram.wiki/use-case/persistent-memory-avatar-second-life https://vram.wiki/use-case/strix-halo-qwen-flash-next-projects https://vram.wiki/use-case/nda-developer-small-local-tasks https://vram.wiki/use-case/radeon-9070xt-light-automation https://vram.wiki/use-case/hermes-overnight-cron-subagent https://vram.wiki/use-case/qwen-coding-3090-5060-hybrid https://vram.wiki/use-case/amd-rocm-three-layer-agent-architecture https://vram.wiki/use-case/qwen-coding-cline-4070-ti-super https://vram.wiki/use-case/autonomous-website-qwen-7900xtx https://vram.wiki/use-case/local-process-orchestration-comfyui-voice https://vram.wiki/use-case/iterative-game-development-pi-64gb-vram https://vram.wiki/use-case/private-transcription-translation-handwriting https://vram.wiki/use-case/strix-halo-qwen-deep-work https://vram.wiki/use-case/dual-r9700-coding-homelab https://vram.wiki/use-case/custom-llamacpp-seven-shipped-projects https://vram.wiki/use-case/privacy-freelance-four-3090-workstation https://vram.wiki/use-case/hermes-document-review-nas-automation https://vram.wiki/use-case/hybrid-rag-m2-ultra-m3-max https://vram.wiki/use-case/structured-coding-m1-max-qwen https://vram.wiki/use-case/local-qwen-personal-coding-comparison https://vram.wiki/use-case/dual-3090-qwen-infrastructure https://vram.wiki/use-case/local-code-review-against-claude https://vram.wiki/use-case/rtx-5090-daily-agent-workflows https://vram.wiki/use-case/hybrid-frontier-local-token-saver https://vram.wiki/use-case/yolo-security-camera-power-optimization https://vram.wiki/use-case/two-server-local-ai-ecosystem https://vram.wiki/use-case/low-vram-agent-experiments-rtx-5060 https://vram.wiki/use-case/qwen-flash-ada-laptop-deepseek-harness https://vram.wiki/use-case/rtx-pro-6000-vllm-workstation https://vram.wiki/use-case/sales-agent-glm-dgx-sparks https://vram.wiki/use-case/legal-anonymize-ocr-rag-local-execution https://vram.wiki/use-case/client-code-nda-unit-tests-3090 https://vram.wiki/use-case/qwen-27b-96k-q8-kv-3090 https://vram.wiki/use-case/comfyui-video-product-model-m1-max https://vram.wiki/use-case/life-automation-lab-monitoring-architect https://vram.wiki/use-case/strix-halo-128gb-research-stack https://vram.wiki/use-case/local-not-worth-it-electricity-cost https://vram.wiki/use-case/sensor-csv-to-report-deterministic https://vram.wiki/use-case/obsidian-calendar-morning-brief https://vram.wiki/use-case/hermes-invoice-emails-m5-pro https://vram.wiki/use-case/local-video-generation-minimax-cost-saving https://vram.wiki/use-case/small-systems-knowledge-base-translation https://vram.wiki/use-case/mac-studio-512gb-omlx-coding https://vram.wiki/use-case/text-adventure-engine-local-checker https://vram.wiki/use-case/business-apps-local-endpoint-9060xt https://vram.wiki/use-case/hermes-finance-tracker-mcp https://vram.wiki/use-case/local-deep-research-gpt-oss-gemma https://vram.wiki/use-case/job-listing-agent-local-cloud-split https://vram.wiki/use-case/ssh-raspberry-pi-camera-setup https://vram.wiki/use-case/office-helpdesk-ops-two-5090-vllm https://vram.wiki/use-case/strix-halo-business-cloud-migration https://vram.wiki/use-case/llamampere-fork-high-throughput https://vram.wiki/use-case/orchestrator-five-subagent-harness https://vram.wiki/use-case/persona-profiles-gpt-oss-swift https://vram.wiki/use-case/synthetic-helpdesk-tickets-3080 https://vram.wiki/use-case/under-3000-dollars-fully-local-server https://vram.wiki/use-case/work-rag-classification-sensitive-data https://vram.wiki/use-case/5090-three-concurrent-sessions https://vram.wiki/use-case/sglang-nvfp4-180k-qwen27b https://vram.wiki/use-case/dual-9070xt-budget-32gb https://vram.wiki/use-case/cpu-only-dedi-gpt-oss-20b https://vram.wiki/use-case/frontier-orchestrates-ninfer-local https://vram.wiki/use-case/solo-lawyer-low-spec-laptops https://vram.wiki/use-case/27b-reasoning-model-on-2x-rtx-2080-ti-nvfp4-vllm-for-agentic-orchestration https://vram.wiki/use-case/math-ocr-pdf-qwen-vision-5070ti https://vram.wiki/use-case/creative-writing-and-a-discord-bot-on-an-8gb-rtx-4060 https://vram.wiki/vram/8gb-or-less https://vram.wiki/vram/10-12gb https://vram.wiki/vram/16gb https://vram.wiki/vram/17-20gb https://vram.wiki/vram/21-24gb https://vram.wiki/vram/25-32gb https://vram.wiki/vram/33-48gb https://vram.wiki/vram/49-96gb https://vram.wiki/vram/97gb-plus https://vram.wiki/vram/12-16gb https://vram.wiki/vram/20gb https://vram.wiki/vram/24gb https://vram.wiki/vram/32-64gb https://vram.wiki/vram/96gb-plus https://vram.wiki/category/coding https://vram.wiki/category/hybrid-orchestration https://vram.wiki/category/homelab-ops https://vram.wiki/category/document-processing https://vram.wiki/category/regulated-work https://vram.wiki/category/personal-assistant https://vram.wiki/category/batch-utility https://vram.wiki/category/media https://vram.wiki/category/business-automation https://vram.wiki/category/research-analysis https://vram.wiki/model/qwen-3-8-27b https://vram.wiki/model/qwen-3-8-flash-next https://vram.wiki/model/qwen-3-6-35b-a3b https://vram.wiki/model/qwen3-coder https://vram.wiki/model/qwen3-vl-4b https://vram.wiki/model/qwen-3-5-27b https://vram.wiki/model/qwen-3-5-9b https://vram.wiki/model/qwen-3-5-122b https://vram.wiki/model/qwen-2-5-7b https://vram.wiki/model/qwen-2-5-coder-7b https://vram.wiki/model/qwen-2-5-3b https://vram.wiki/model/qwen https://vram.wiki/model/gemma-4 https://vram.wiki/model/gemma-4-8b https://vram.wiki/model/gemma-4-12b https://vram.wiki/model/gemma-4-26b https://vram.wiki/model/gemma-4-31b https://vram.wiki/model/deepseek-v4-flash https://vram.wiki/model/deepseek-v4-flash-vision https://vram.wiki/model/kimi-k3 https://vram.wiki/model/granite-4-1 https://vram.wiki/model/gpt-oss-120b https://vram.wiki/model/devstral https://vram.wiki/model/muse-glimmer https://vram.wiki/model/mimo-2-5 https://vram.wiki/model/ling-3-flash https://vram.wiki/model/tiel-coder-35b-a3b-mlx https://vram.wiki/model/ornith-1-5-9b https://vram.wiki/model/qwen-3-8-9b https://vram.wiki/model/qwen-3-5-35b https://vram.wiki/model/glm-5-3-flash https://vram.wiki/model/gpt-oss-20b https://vram.wiki/runtime/ollama https://vram.wiki/runtime/llama-cpp https://vram.wiki/runtime/lm-studio https://vram.wiki/runtime/llama-cpp-python https://vram.wiki/runtime/text-generation-webui https://vram.wiki/runtime/lmdeploy https://vram.wiki/runtime/exllamav2 https://vram.wiki/runtime/localai https://vram.wiki/runtime/koboldcpp https://vram.wiki/runtime/mlx https://vram.wiki/runtime/openvino-genai https://vram.wiki/runtime/vllm https://vram.wiki/runtime/sglang https://vram.wiki/runtime/text-generation-inference https://vram.wiki/runtime/tensorrt-llm https://vram.wiki/runtime/tabbyapi https://vram.wiki/runtime/onnx-runtime https://vram.wiki/runtime/mlc-llm https://vram.wiki/runtime/transformers https://vram.wiki/runtime/llamafile https://vram.wiki/runtime/jan https://vram.wiki/runtime/gpt4all https://vram.wiki/runtime/ik-llama-cpp https://vram.wiki/runtime/ktransformers https://vram.wiki/runtime/ramalama https://vram.wiki/runtime/llama-swap https://vram.wiki/runtime/mistral-rs https://vram.wiki/runtime/comfyui https://vram.wiki/runtime/automatic1111-webui https://vram.wiki/runtime/invokeai https://vram.wiki/runtime/diffusers https://vram.wiki/runtime/sentence-transformers https://vram.wiki/runtime/whisper-cpp https://vram.wiki/runtime/faster-whisper https://vram.wiki/runtime/deepseek-harness https://vram.wiki/runtime/ninfer https://vram.wiki/runtime/halogen https://vram.wiki/runtime/lucebox https://vram.wiki/runtime/omlx https://vram.wiki/runtime/open-webui https://vram.wiki/runtime/pi https://vram.wiki/gpu/rtx-3060 https://vram.wiki/gpu/rtx-3070 https://vram.wiki/gpu/rtx-3070-ti https://vram.wiki/gpu/rtx-3080 https://vram.wiki/gpu/rtx-3080-ti https://vram.wiki/gpu/rtx-3090 https://vram.wiki/gpu/rtx-4070 https://vram.wiki/gpu/rtx-4070-ti-super https://vram.wiki/gpu/rtx-4090 https://vram.wiki/gpu/rtx-4000-sff-ada https://vram.wiki/gpu/rtx-5070 https://vram.wiki/gpu/rtx-5070-ti https://vram.wiki/gpu/rtx-5090 https://vram.wiki/gpu/rtx-5060 https://vram.wiki/gpu/rtx-5080 https://vram.wiki/gpu/rtx-2070-ti-super https://vram.wiki/gpu/rtx-6000-pro https://vram.wiki/gpu/a100-40gb https://vram.wiki/gpu/a100-80gb https://vram.wiki/gpu/h200 https://vram.wiki/gpu/b200 https://vram.wiki/gpu/nvidia-p100 https://vram.wiki/gpu/dgx-spark https://vram.wiki/gpu/radeon-6800xt https://vram.wiki/gpu/radeon-rx-6650-xt https://vram.wiki/gpu/radeon-7900xt https://vram.wiki/gpu/radeon-7900xtx https://vram.wiki/gpu/radeon-7800-xt https://vram.wiki/gpu/radeon-pro-ai-r9700 https://vram.wiki/gpu/amd-v620 https://vram.wiki/gpu/radeon-9070-xt https://vram.wiki/gpu/radeon-780m https://vram.wiki/gpu/strix-halo https://vram.wiki/gpu/intel-arc-a770m https://vram.wiki/gpu/nuc12 https://vram.wiki/gpu/mac-m1-max https://vram.wiki/gpu/mac-m2-ultra https://vram.wiki/gpu/mac-m3-max https://vram.wiki/gpu/mac-m4-max https://vram.wiki/gpu/mac-m5 https://vram.wiki/gpu/mac-m5-max https://vram.wiki/gpu/mac-mini https://vram.wiki/gpu/mac-m3-ultra https://vram.wiki/gpu/mac-m4-pro https://vram.wiki/gpu/mac-m5-pro https://vram.wiki/gpu/rtx-4060 https://vram.wiki/gpu/rtx-5060-ti https://vram.wiki/gpu/gtx-1650 https://vram.wiki/gpu/radeon-9060-xt https://vram.wiki/gpu/nvidia-jetson https://vram.wiki/gpu/rtx-2060 https://vram.wiki/gpu/rtx-3050 https://vram.wiki/gpu/rtx-3060-ti https://vram.wiki/gpu/rtx-3090-ti https://vram.wiki/gpu/rtx-4060-ti https://vram.wiki/gpu/rtx-4070-super https://vram.wiki/gpu/rtx-4080-super https://vram.wiki/gpu/rtx-4070-laptop https://vram.wiki/gpu/rtx-4090-laptop https://vram.wiki/gpu/rtx-2000-ada https://vram.wiki/gpu/rtx-4500-ada https://vram.wiki/gpu/rtx-5000-ada https://vram.wiki/gpu/rtx-6000-ada https://vram.wiki/gpu/rtx-pro-4000-blackwell https://vram.wiki/gpu/rtx-pro-4500-blackwell https://vram.wiki/gpu/rtx-pro-5000-blackwell https://vram.wiki/gpu/rtx-pro-6000-blackwell-workstation-edition https://vram.wiki/gpu/rtx-pro-6000-blackwell-server-edition https://vram.wiki/gpu/nvidia-a10 https://vram.wiki/gpu/nvidia-a30 https://vram.wiki/gpu/nvidia-a40 https://vram.wiki/gpu/nvidia-l4 https://vram.wiki/gpu/nvidia-l40s https://vram.wiki/gpu/nvidia-a100 https://vram.wiki/gpu/nvidia-h100 https://vram.wiki/gpu/nvidia-h200 https://vram.wiki/gpu/nvidia-b200 https://vram.wiki/gpu/rx-6700-xt https://vram.wiki/gpu/rx-6800 https://vram.wiki/gpu/rx-6900-xt https://vram.wiki/gpu/rx-7600-xt https://vram.wiki/gpu/rx-7700-xt https://vram.wiki/gpu/rx-9070 https://vram.wiki/gpu/radeon-pro-w7600 https://vram.wiki/gpu/radeon-pro-w7700 https://vram.wiki/gpu/radeon-ai-pro-r9700 https://vram.wiki/gpu/amd-instinct-mi325x https://vram.wiki/gpu/amd-instinct-mi350x https://vram.wiki/gpu/radeon-pro-w7800 https://vram.wiki/gpu/radeon-pro-w7900 https://vram.wiki/gpu/amd-instinct-mi210 https://vram.wiki/gpu/amd-instinct-mi300x https://vram.wiki/gpu/intel-arc-a750 https://vram.wiki/gpu/intel-arc-a770 https://vram.wiki/gpu/intel-arc-b570 https://vram.wiki/gpu/intel-arc-b580 https://vram.wiki/gpu/intel-arc-pro-b50 https://vram.wiki/gpu/intel-arc-pro-b60 https://vram.wiki/gpu/rtx-4060-laptop https://vram.wiki/gpu/rtx-4050-laptop https://vram.wiki/gpu/rtx-4080-laptop https://vram.wiki/gpu/rtx-4000-ada-laptop https://vram.wiki/gpu/rtx-5000-ada-laptop https://vram.wiki/gpu/rtx-5090-laptop https://vram.wiki/gpu/rtx-5080-laptop https://vram.wiki/gpu/rtx-a2000 https://vram.wiki/gpu/rtx-a4000 https://vram.wiki/gpu/rtx-a5000 https://vram.wiki/gpu/rtx-a6000 https://vram.wiki/gpu/nvidia-t4 https://vram.wiki/gpu/nvidia-v100 https://vram.wiki/gpu/nvidia-l20 https://vram.wiki/gpu/nvidia-gh200 https://vram.wiki/gpu/nvidia-gb200 https://vram.wiki/gpu/radeon-pro-w6800 https://vram.wiki/gpu/radeon-rx-7900m https://vram.wiki/gpu/amd-instinct-mi250x https://vram.wiki/gpu/amd-instinct-mi300a https://vram.wiki/gpu/intel-arc-pro-a60 https://vram.wiki/gpu/intel-data-center-gpu-max-1550 https://vram.wiki/gpu/intel-arc-a730m https://vram.wiki/gpu/rtx-2000-ada-laptop https://vram.wiki/gpu/rtx-3000-ada-laptop https://vram.wiki/gpu/rtx-3500-ada-laptop https://vram.wiki/gpu/rtx-6000-ada-laptop https://vram.wiki/gpu/nvidia-a2 https://vram.wiki/gpu/nvidia-a16 https://vram.wiki/gpu/nvidia-l40 https://vram.wiki/gpu/nvidia-a800 https://vram.wiki/gpu/nvidia-h800 https://vram.wiki/gpu/nvidia-h20 https://vram.wiki/gpu/radeon-rx-7600m https://vram.wiki/gpu/radeon-rx-7700s https://vram.wiki/gpu/radeon-rx-7800m https://vram.wiki/gpu/radeon-pro-w7500 https://vram.wiki/gpu/amd-instinct-mi100 https://vram.wiki/gpu/amd-instinct-mi250 https://vram.wiki/gpu/intel-arc-a370m https://vram.wiki/gpu/intel-arc-a550m https://vram.wiki/gpu/intel-arc-pro-a40 https://vram.wiki/gpu/gtx-1080-ti https://vram.wiki/gpu/rtx-2070-super https://vram.wiki/gpu/rtx-2080-ti https://vram.wiki/gpu/quadro-rtx-6000 https://vram.wiki/gpu/quadro-rtx-8000 https://vram.wiki/gpu/tesla-p40 https://vram.wiki/gpu/nvidia-h100-pcie https://vram.wiki/gpu/nvidia-h200-pcie https://vram.wiki/gpu/rx-6600 https://vram.wiki/gpu/rx-6600-xt https://vram.wiki/gpu/rx-6750-xt https://vram.wiki/gpu/rx-6950-xt https://vram.wiki/gpu/radeon-rx-6850m-xt https://vram.wiki/gpu/amd-instinct-mi60 https://vram.wiki/gpu/intel-arc-a380 https://vram.wiki/gpu/intel-arc-a580 https://vram.wiki/gpu/rtx-a5500 https://vram.wiki/gpu/rtx-a4500 https://vram.wiki/gpu/b100 https://vram.wiki/gpu/h100-nvl https://vram.wiki/gpu/rx-7600 https://vram.wiki/gpu/rx-9060 https://vram.wiki/gpu/gaudi-3 https://vram.wiki/gpu/gtx-1060 https://vram.wiki/gpu/gtx-1070 https://vram.wiki/gpu/gtx-1070-ti https://vram.wiki/gpu/gtx-1660-super https://vram.wiki/gpu/gtx-1660-ti https://vram.wiki/gpu/rtx-2060-super https://vram.wiki/gpu/rtx-2070 https://vram.wiki/gpu/rtx-3050-ti-laptop https://vram.wiki/gpu/rx-580 https://vram.wiki/gpu/rx-590 https://vram.wiki/gpu/rx-5700-xt https://vram.wiki/gpu/rx-5600-xt https://vram.wiki/gpu/radeon-pro-w7900-ds https://vram.wiki/gpu/gtx-1080 https://vram.wiki/gpu/rtx-5050