4090 vs 5090 for 70B quants with TensorRT-LLM

Thread Rating:

0 Vote(s) - 0 Average
1
2
3
4
5

Linear Mode

4090 vs 5090 for 70B quants with TensorRT-LLM

rocmriley Offline

Member

Posts: 211
Threads: 21
Joined: Sep 2026
Reputation: 0

10-18-2026, 03:12 AM

Tried to reproduce it just now. My result was mixed: first run looked bad, second run looked normal. The surprise was that memory pressure showed up before throughput collapsed. At this point the real bottleneck is my patience.

7900 XTX / ROCm / somehow still optimistic

Find

« Next Oldest | Next Newest »

Messages In This Thread

4090 vs 5090 for 70B quants with TensorRT-LLM - by turbogrid11 - 10-17-2026, 10:35 PM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by vrambandit - 10-17-2026, 10:51 PM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by rocmriley - 10-18-2026, 03:12 AM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by cudapoint - 10-18-2026, 03:54 AM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by llmtrack - 10-18-2026, 08:32 PM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by vrambandit - 10-18-2026, 07:03 PM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by embersmith - 10-18-2026, 07:45 PM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by batchwire - 10-18-2026, 07:56 PM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by routecrate - 10-18-2026, 08:08 PM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by amberhall - 10-19-2026, 02:37 AM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by torchshed - 10-19-2026, 03:27 AM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by fusedhall56 - 10-19-2026, 04:00 AM

Re: 4090 vs 5090 for 70B quants with TensorRT-LLM - by batchwire - 10-19-2026, 05:27 AM

View a Printable Version

Forum Jump:

Users browsing this thread: 1 Guest(s)

Login
Username:
Password:	Lost Password?
	Remember me

Sign in with X.com