Builds / RTX 5090
RTX 5090 workstation: Qwen3.8 Flash-Next IQ3_S at 159.2 tok/s on Strata (expert offload)
Source: akitaonrails.github.io ↗📏 How measured: reported by the original source; RigBase re-checked the numbers at that source
- GPU(s)
- RTX 5090
Amazon · eBay (used) · Newegg - VRAM / unified memory (GB)
- 32
- CPU
- not stated
- Motherboard
- not stated
- RAM
- not stated
- PSU
- not stated
- Other
- Strata expert-offload engine: hot experts on the GPU, all experts in system RAM, ~29 GB lookup table on SSD; GPU also drives the Hyprland desktop (4 GiB reserve)
- Reported total cost
- not stated
- Power
- not stated
Measured speed (generation tokens/sec)
| Model | Quant | Engine | tok/s | Details |
|---|---|---|---|---|
| Qwen3.8 Flash-Next | IQ3_S (~55 GB), experts offloaded to RAM, speculation helper | Strata | 159.2 |
AkitaOnRails LLM Benchmark v4, 22 Oct 2026. Qwen3.8 Flash-Next runs with CPU offload of experts (Strata). Scored 91.0 (Tier A) on the author's code-review benchmark. With the 4 GiB desktop reserve, throughput in the initial measurement went from ~200 to ~150 tok/s. Ornith-1.5-35B-A3B Q8_0 on the same card/engine: ~80 tok/s.
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.
- 1× RTX 5090: ≈€7,140 each (≈$8,003), new, approx. price as of 2026-10-24, lowest listing on billiger.de (MSI GeForce RTX 5090 (lowest variant))
New retail prices in Germany incl. VAT; used cards are often much cheaper. Prices change daily.
Source: https://akitaonrails.github.io/en/2026/10/22/llm-benchmark-v4-consumer-open-source-rtx5090-strix-halo-strata/
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 RTX 5090 run Qwen3.8 Flash-Next?