Inference › LLM Models › Proprietary / Frontier v
70B still worth the pain

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Threaded Mode
70B still worth the pain
vramghost Offline
Junior Member
**
Posts: 41
Threads: 5
Joined: Aug 2026
Reputation: 0
#11
10-19-2026, 03:35 AM
Did a small comparison this afternoon. On 5090 with MLX, I saw basically the same behavior. The surprise was that memory pressure showed up before throughput collapsed.
Find
Reply
tinycontext Offline
Member
***
Posts: 112
Threads: 12
Joined: Sep 2026
Reputation: 0
#12
10-19-2026, 05:51 AM
that fixed it for me too.
context windows are not a product strategy
Find
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)