-
ENV:
cuda: 12.1
torch: 2.5.0+cu121
python benchmark_fp6.py
Hello, have you tested the performance of the …
-
The delete device button is currently very large and prominent on the devices page. Hiding the button in a 3 dots menu, or adding a speed bump will prevent accidental deletions.
-
If you apply from a large remote repo, like https://github.com/e2b-dev/e2b-cookbook/blob/main/.grit/patterns/e2b_v0_to_v1_js.md, it is pretty slow since we currently do a full clone.
We should swit…
-
The program is extremely slow, I am measuring 51.81MB/s speed only on Windows with 4 providers from 4 different backbones.
@Tensai75 can you please improve the programs performance, so it would work …
-
Hi @frdel, thanks for your great work!
I found AutoMemory to be quite inefficient because it produces a summarization of the same memories at every step of the conversation.
### Suggestion
My…
-
Hello,
As we have tested telemetry update speed at 2000,16 which is around 125 fps it is too fast for telemetry hardware based devices like Cammus hubs for RPM LED's and Speed/Gear LCD it causes de…
-
Hi this is the same model loaded in ThreeJS (chrome)
https://github.com/user-attachments/assets/de81d590-3af3-4171-8d8b-ac6cc0024c0f
and this in windows forms opentk control
https://githu…
merco updated
4 weeks ago
-
Me, and many people are upset at how the new AI is unfortunately, extremely buggy, and sometimes even frustrating to play against. Objectively, the old AI was miles better than the one we have current…
-
Hi,
Looking at running various models with various inputs - it seems a lot of time for the initial runs is being spent benchmarking potential kernels - including the naive ones (e.g. `naive_conv_n…
-
First off, this is a wonderful script and I 100% understand you probably can't do this. But the speed of everything is so very slow. I'm not sure if there's a different way you could parse the playlis…