#5680 [Benchmark] 2026-11-09-community-rtx-5070-plus-rtx-5060-ti-qwen3-8-flash-next-iq3-s

closed · @abelbranislav · 3 comments · View on GitHub

BenchmarksSetup & installAMD / HIPNVIDIA / CUDAModels & quantsDocumentationWindows

Description

Benchmark results: [README.md](https://github.com/user-attachments/files/34595801/README.md)

Hardware and software:
GPU and VRAM: NVIDIA GeForce RTX 5070 + NVIDIA GeForce RTX 5060 Ti, 28 GiB VRAM
CPU: AMD Ryzen 7 5800XT 8-Core Processor (8 cores, 16 threads); installed RAM: 63.9 GiB; storage: ST2000DL004 HD204UI HDD/SATA; SanDisk SDSSDH3 2T00 SSD/SATA; SanDisk SDSSDH3 4T00 SSD/SATA; Samsung SSD 980 PRO with Heatsink 2TB SSD/NVMe
OS: Windows 10 (10.0.19045); GPU driver: 621.14; CUDA (driver supports up to): 13.5; engine built for CUDA 13.0

Strata: project 0.1.59, commit f33c8c30; engine 0.1.55
GPU detail (after the run, near idle): power_limit_w=250.00, power_draw_w=87.71, pcie=gen4 x16, pstate=P1, temp_c=53.04
GPU detail (after the run, near idle): power_limit_w=180.00, power_draw_w=51.84, pcie=gen4 x4, pstate=P1, temp_c=49.04
Background workloads: not recorded (GPU power limit/draw read above)

Model and configuration
Model and quantization: qwen3.8-flash-next-iq3_s
Context: 131072 tokens; parallel slots: 1
Settings: the install's own configuration, unchanged; thinking off for the suite; temperature 0

Related on this site

Quick links to related pages.