Upgrading to dual Radeon AI PRO R9700 (64GB VRAM) for Qwen 3.8 27B local inference
Athabasco · reddit · 2026-10-05
- The author currently runs one ASRock Radeon AI PRO R9700 (32GB) with 64GB RAM and plans to add a second R9700 (CA$2,499.99) with x8/x8 bifurcation to run Qwen3.8-Flash-Next 3.8 27B at Q4 or higher quants.
- Full build: Ryzen 9 7900X3D, Gigabyte B850 AI TOP, 64GB DDR5-6000 CL30, 1200W Platinum PSU; total CA$6,307.84. Soliciting feedback before buying remaining parts.
More from Infra
- Only 2% of US electricians are DC-certified, says Ben Horowitz — AI infra bottleneck — rohanpaul_ai · 2026-10-05
- OpenRouter routing distorts inference pricing: GLM 5.3 at $0.08/M input vs $5/M output — soumithchintala · 2026-10-05
- United Airlines to equip 880+ aircraft with Starlink by end of 2026 — elonmusk · 2026-10-05
- Nutanix Buys Ryax to Squeeze More Work Out of Enterprises' Idle GPUs — shashib · 2026-10-05
- Neoclouds locking 3-year GPU leases are 'downright stupid' as MI300X prices rebound from $1.50 to $3.39 — AccBalanced · 2026-10-05
- llama.cpp now supports Clef, OpenJev and other decision models — ngxson · 2026-10-05