# What did I miss? What should I be running on 2x16Gb VRAM and 128Gb RAM? — retained record

> Retained reddit record from reddit:LocalLLaMA.

Canonical: https://highsignal.app/data/records/98462618565b86f1

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:LocalLLaMA)  published 2026-10-31 19:00 ingested 2026-11-01 02:40

[original source ↗](https://www.reddit.com/r/LocalLLaMA/comments/1xkieqt/what_did_i_miss_what_should_i_be_running_on/)

## retained text



```
tl;dr: What should I be running for coding/agentic workflows? I have 2xRTX 5080s and 128Gb DDR5 RAM. 

Currently, I'm running:   
  
\-Qwen3.8-Flash-Next-Q5\_K\_XL (262k ctx, pp\~150-200tps, decode \~15tps)  
  
\-Qwen3.8-27B-exl3-6bpw-H6 (131k ctx, pp \~2k-3ktps, decode \~60-80tps)

\-DeepSeek-V4-Flash-exl3-3.04bpw (1M ctx, pp\~100-900, decode \~15-20tps)

I've been in a cave hiding from my local model optimization addiction for a few weeks. What did I miss? Should I be using Strata? Is llama.cpp old news now? Any new models worth trying?
```



document fetched 2026-11-01 00:38 · canonical https://reddit.com/r/LocalLLaMA/comments/1xkieqt/what_did_i_miss_what_should_i_be_running_on
