bench: community results for nvidia-geforce-rtx-5060-ti

[AythaE]

Automated benchmark contribution from llmfit bench --share.

Model	Provider	Avg TPS	Avg TTFT (ms)
mattbucci/gemma-4-12B-AWQ	vllm	47.2	—

Submitted without the gh CLI via the GitHub device flow.

[greptile-apps]

Confidence Score: 5/5

[Low impact] The PR appears safe to merge.

Summary

Adds two community benchmark results for an NVIDIA GeForce RTX 5060 Ti:

• Two models now have benchmark results for the RTX 5060 Ti system.

Both report three runs through vllm. Since the previous review, only the second submission was added. No actionable issues were found.

_{Reviews (2) · Last reviewed commit: "data: community benchmark (nvidia-geforc..." · Reviewed by Greptile}

[AlexsJones]

Thanks for sharing these, @AythaE, vLLM AWQ numbers on a 5060 Ti are a welcome addition since there's no folder for that card yet.

Checked both files: same machine in each (Ryzen 7 3700X, 30.26 GB RAM, RTX 5060 Ti with 15.93 GB VRAM, classed DISCRETE_GPU), one model per file with three runs each and a tight min/max spread, written by llmfit 1.1.17 against schema v1. The second file (MiMo-V2.6-Distill-Qwen-9B at 30.9 tok/s) isn't in the PR description's table, but it's from the same session, so no need to change anything.

Looks good to me. Only the data-only checks ran here (no Rust test run for a data path), so I'll leave the merge to Alex.