|
when TensorRT-LLM clearly wins with MLX
|
10-15-2026, 03:31 AM
Quote:My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Llama and compare from there. Good point. I don't think your conclusion is far off. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.
Mostly CUDA, sometimes wrong, usually measured
|
|
« Next Oldest | Next Newest »
|
| Messages In This Thread |
|
when TensorRT-LLM clearly wins with MLX - by rasterbench - 10-13-2026, 09:31 PM
Re: when TensorRT-LLM clearly wins with MLX - by pixelcrate82 - 10-14-2026, 12:13 AM
Re: when TensorRT-LLM clearly wins with MLX - by northcraft90 - 10-14-2026, 04:21 AM
Re: when TensorRT-LLM clearly wins with MLX - by macrotinker - 10-14-2026, 06:47 AM
Re: when TensorRT-LLM clearly wins with MLX - by kernelmike - 10-14-2026, 02:51 PM
Re: when TensorRT-LLM clearly wins with MLX - by syshouse - 10-15-2026, 02:52 AM
Re: when TensorRT-LLM clearly wins with MLX - by kernelmike - 10-15-2026, 03:31 AM
|
