Builds / RTX 5070 Ti
RTX 5070 Ti + Ryzen 9 5950X: gpt-oss 20B at 213.2 tok/s, Qwen3.5 9B at 126.6 (llama.cpp CUDA)
Last updated:
- GPU(s)
- RTX 5070 Ti
Amazon · eBay (used) · Newegg - VRAM / unified memory (GB)
- 16
- CPU
- AMD Ryzen 9 5950X (16 cores / 32 threads)
Amazon · eBay (used) · Newegg - Motherboard
- Gigabyte B550 AORUS PRO AC
Amazon · eBay (used) · Newegg - RAM
- 32 GB DDR4-4000 (2x16 GB G.Skill F4-4000C16-16GTZRA)
Amazon · eBay (used) · Newegg - PSU
- not stated
- Other
- Windows 11 Pro; llama.cpp b11669 CUDA
- Reported total cost
- not stated
- Power
- not stated
Measured speed (generation tokens/sec)
| Model | Quant | Engine | tok/s | Prompt processing (tok/s) | Details |
|---|---|---|---|---|---|
| gpt-oss 20B | MXFP4 (12.1 GB GGUF), reasoning low | llama.cpp b11669 (CUDA) | 213.2 | 7,884 @ 15,000-token prompt est. TTFT @8k ≈ 1.0 s (calculated from measured prefill) | context: code prompt, 32K window📏 measuredlocally test: one user, thinking off, same code prompt on every machine; middle of three runs (idle machine, 8 quiet seconds before each run); llama.cpp b11669, 32,768-token window; whole model on the card; run with --skip-chat-parsing |
| Qwen3.5 9B | Q4_K_M (5.7 GB) | llama.cpp b11669 (CUDA) | 126.6 | 5,275 @ 19,000-token prompt est. TTFT @8k ≈ 1.6 s (calculated from measured prefill) | context: code prompt, 32K window📏 measuredlocally test: one user, thinking off, same code prompt on every machine; middle of three runs (idle machine, 8 quiet seconds before each run); llama.cpp b11669, 32,768-token window; whole model on the card |
| Gemma 4 12B | QAT Q4_0 (7.0 GB, Google's file) | llama.cpp b11669 (CUDA) | 92.5 | 2,910 @ 19,000-token prompt est. TTFT @8k ≈ 2.8 s (calculated from measured prefill) | context: code prompt, 32K window📏 measuredlocally test: one user, thinking off, same code prompt on every machine; middle of three runs (idle machine, 8 quiet seconds before each run); llama.cpp b11669, 32,768-token window; whole model on the card |
| Gemma 4 26B-A4B | UD-IQ4_XS (13.6 GB) | llama.cpp b11669 (CUDA), partial CPU offload | 59.4 | 2,199 @ 19,000-token prompt est. TTFT @8k ≈ 3.7 s (calculated from measured prefill) | context: code prompt, 32K window📏 measuredlocally test: one user, thinking off, same code prompt on every machine; middle of three runs (idle machine, 8 quiet seconds before each run); llama.cpp b11669, 32,768-token window; experts of 15 of 30 layers kept in system RAM (llama.cpp's fit tool chose the split) |
Prompt processing = prefill speed (how fast the prompt is read) as reported by the source, with the prompt length it was measured at; blank = the source does not report it. est. TTFT @8k is calculated from the measured prefill speed (8,192 ÷ prefill tok/s), shown only when prefill was measured on a prompt of 512+ tokens. It is not a measurement and ignores load time and caching. measured TTFT is the source's own time-to-first-token. Why prefill matters for coding assistants.
measuredlocally.com (Alex, UK), tested 20-21 Oct 2026, full parts list published. A 16 GB card runs 9B-20B models fully on the GPU; Gemma 4 26B-A4B needs half its experts in system RAM and still decodes at 59 tok/s. A ~1.75-bit Ternary Bonsai run is below Q4 and not listed.
- 1× RTX 5070 Ti: ≈€1,245 each (≈$1,395), new, approx. price as of 2026-10-24, lowest listing on billiger.de (PNY GeForce RTX 5070 Ti 16 GB (listed as "GDDR6"))
New retail prices in Germany incl. VAT; used cards are often much cheaper. Prices change daily.
https://measuredlocally.com/hardware-leaderboard.html
Blank fields = not stated in the source. We never estimate build data; GPU street prices above are separately sourced and dated.
Copy into a README or blog. Links back to this build. See badge docs (CC BY 4.0 attribution). The SVG does not invent speeds — numbers stay on the build page.
Markdown
[](https://getrigbase.com/builds/rtx-5070-ti-ryzen-9-5950x-32gb-measuredlocally/)
HTML
<a href="https://getrigbase.com/builds/rtx-5070-ti-ryzen-9-5950x-32gb-measuredlocally/" rel="noopener"><img src="https://getrigbase.com/badge/rtx-5070-ti-ryzen-9-5950x-32gb-measuredlocally.svg" alt="Measured on RigBase" height="20"/></a>
Preview: 
· How to use
Quick answers: Can RTX 5070 Ti run Gemma 4 12B? · Can RTX 5070 Ti run Gemma 4 26B-A4B? · Can RTX 5070 Ti run Qwen3.5 9B? · Can RTX 5070 Ti run gpt-oss 20B?