Life after an RTX 3090 > Life before an RTX 3090 Buy a GPU
@theahmadosman
-
Llama.cpp Alternatives Recommended for Multi-GPU Inference Setups
By
–
Keep in mind that once you get into concurrency and tensor parallelism (multi-GPU setups), you should be recommending inference engines other than llama.cpp
-
Incredible Resource to Learn How LLMs Work and Evolved
By
–
INCREDIBLE RESOURCE to learn how LLMs work and how they evolved overtime
-
Use MoE Models for Unified Memory Hardware Like DGX Spark
By
–
As I have mentioned before, stop trying to get Dense models running on the DGX Spark/Mac Studios Unified Memory is best fit for MoEs because you only make each token go through a small subset of the numbers of parameters in the model Optimize for your hardware x.com/LeTechLead/sta…
-
Qwen 3.5 27B Dense vs 122B MoE Compared on DGX Spark
By
–
This guy has written an article comparing > Qwen 3.5 27B Dense vs > Qwen 3.5 122B MoE on the DGX Spark
-
Qwen 3.5 27B Matches Sonnet 4.5 on Single RTX 5090
By
–
That model, Qwen 3.5 27B Dense is equal to Sonnet 4.5 Runs great on a single RTX 5090 w/ full context But we are not anywhere near Opus 4.5 even with Qwen 3.5 397B MoE
-
Kimi K2.5 First End-to-End Run in NVFP4 Precision
By
–
Not saying that OP isn’t farming. But Kimi K2.5 was only post-trained in NVFP4. This is the first end-to-end run in NVFP4.
-
Full Model GPU Offload Required to Avoid Performance Bottlenecks
By
–
Partial offloads wouldn’t work because this will create a severe bottleneck That’s why the full model has to be offloaded to the GPU to see actual performance gains
-
GPU Offloading Offers 1800Gbps vs DGX Spark’s 273Gbps
By
–
Yeah that is definitely faster because you’re fully offloading to the GPU which runs at ~1800Gbps in comparison to DGX Spark’s 273Gbps
-
New Website to Benchmark Hardware Tokens per Second Performance
By
–
Update: @AlpinDale and I agreed to collab on a website that does more accurate calculations for your hardware tokens/sec performance Expect more on this after I am done with GTC this week