toktapethe record, not a recording of it

midagedev

https://github.com/midagedev

Narrowed toquant: q3_k_m ×gpu: rtx-a6000 ×sessions: 4 ×Clear all

3 runs, newest first

DeepSeek-V4.1-Flash Q3_K_M · 4 streams 27.4 tok/s

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

llama-serverb96Q3_K_Mrtx-a6000rtx-3090 linux4 streams! 1 caveat

2026-09-18

DeepSeek-V2-Lite-Chat Q3_K_M · 4 streams 211 tok/s

DeepSeek-V2-Lite-Chat.Q3_K_M.gguf

ik_llama.cppc10fbbccQ3_K_Mrtx-a6000linux4 streams! 1 caveat

2026-09-18

Qwen2.5-7B-Instruct Q3_K_M · 4 streams · dense control 196 tok/s

Qwen2.5-7B-Instruct-Q3_K_M.gguf

ik_llama.cppc10fbbccQ3_K_Mrtx-a6000linux4 streams! 1 caveat

2026-09-18