[Update Benchmark] Basalt's new PRs (Strata fork) new record IQ3_S Qwen3.8 Flash Next on single RTX 5080 and 51 GB RAM with 5500 tok/s prompt processing and up to 275 tok/s peak generation (during opencode coding live session)
retained text
Code and tool calls can peak up to 275 tok/s - Reasoning traces up to 190 tok/s Hardware: 5080 16GB PCIe gen5x16, 64 DDR5 6400, Core Ultra 9 285K [https://github.com/jesdga95/basalt](https://github.com/jesdga95/basalt) Some PRs for the optimizations under my name are still pending (xyehya) but upstream already have solid gains & memory optimizations. and yes the IQ3\_S is super usable.
document fetched 2026-10-30 00:45 · canonical https://reddit.com/r/LocalLLaMA/comments/1xi9sfl/update_benchmark_basalts_new_prs_strata_fork_new