Builds / 2x RTX 5080 16 GB
Dual RTX 5080 on EPYC 7F52 (capy-flash user report)
Source: github.com ↗More source: github.com ↗📏 How measured: reported by the original source; RigBase re-checked the numbers at that source
- GPU(s)
- 2x RTX 5080 16 GB
Amazon · eBay (used) · Newegg - VRAM / unified memory (GB)
- 32
- CPU
- AMD EPYC 7F52 (16-core Zen 2)
Amazon · eBay (used) · Newegg - Motherboard
- not stated
- RAM
- 256 GB DDR4-3200 ECC (8x32 GB)
Amazon · eBay (used) · Newegg - PSU
- not stated
- Other
- PCIe 4.0 x16 each, P2P-patched driver; WD_BLACK SN8100 1 TB
- Reported total cost
- not stated
- Power
- not stated
Measured speed (generation tokens/sec)
| Model | Quant | Engine | tok/s | Details |
|---|---|---|---|---|
| Qwen3.8 Flash-Next | NVFP4 | capy-flash v0.7.2 (CUDA 13.4), agentic decode | 141.3 | 📏 User-reported (issue #11), listed in the capy-flash README |
| Qwen3.8 Flash-Next | NVFP4 | capy-flash v0.7.2, --expert-cpu-threads 15, agentic decode | 144.4 | 📏 User-reported (issue #11), listed in the capy-flash README |
Hardware from user Hotcooler's issue #11 (2026-10-19); speeds from the 'Reported by users' table in the capy-flash README. On a Strata NVFP4 fork the same user reports ~145 tok/s agentic, 110-130 prose. Found via localmaxxing (discovery only); numbers re-checked at the original source on 2026-10-24.
Current GPU street price (estimate, not the builder's cost)
New retail prices in Germany incl. VAT; used cards are often much cheaper. Prices change daily.
- 2× RTX 5080: ≈€1,770 each (≈$1,984), new, approx. price as of 2026-10-24, lowest listing on billiger.de (MSI GeForce RTX 5080 Inspire 3X OC Black 16 GB GDDR7)
New retail prices in Germany incl. VAT; used cards are often much cheaper. Prices change daily.
Source: https://github.com/ValerioDolci/capy-flash/issues/11
https://github.com/ValerioDolci/capy-flash
https://github.com/ValerioDolci/capy-flash
Blank fields = not stated in the source. We never estimate build data; GPU street prices above are separately sourced and dated.
Rent before you buy: try this model on a cloud GPU for a few dollars first: Runpod (referral link; sign up with Google to get the new-user bonus)
Quick answers: Can 2x RTX 5080 run Qwen3.8 Flash-Next?