toktapethe record, not a recording of it

NVIDIA GeForce RTX 3090

79 tapesbloomery51llama-server17ik_llama.cpp7 openai3 exllamav31 qwen3.6-35b-a3b10 qwen2.5-7b-instruct1 qwen3-coder-next-512x2.5b1

Narrowed toquant: q3_k_m ×engine: llama-server ×sessions: 1 ×Clear all

1 run, newest first

DeepSeek-V4.1-Flash Q3_K_M · 1 stream · 347 GB model on 72 GB of VRAM 23.8 tok/sprefill 52.3 tok/s

DeepSeek-V4.1-Flash-Q3_K_M-00001-of-00009.gguf

llama-serverb102Q3_K_M748B · 18B activeP6509 · G66 · cache 0%ctx 8koffload partialrtx-3090rtx-a6000 linux1 stream! 4 caveats

midagedev

2026-09-20

Published runs · every tape on this page was recorded on this card and carries its own caveats; the page does not rank them. toktape publish <run.toktape> puts one here.

Page views are counted with Cloudflare Web Analytics: no cookies, no fingerprinting. The toktape binary itself sends nothing anywhere.