fmperf-project fmperf issues

fmperf-project / fmperf

Cloud Native Benchmarking of Foundation Models

Apache License 2.0

20 stars 10 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Documentations for Open Access

#36 GhaziSyed closed 4 days ago
0
Resolved issue with environment variables

#35 GhaziSyed closed 6 days ago
0
generator function fix

#34 GhaziSyed closed 6 days ago
0
Fix paths that are broken when sweep is executed

#33 maxdebayser closed 1 week ago
1
Make the base directory for requests configurable

#32 maxdebayser closed 1 month ago
1
Move the imports of the tgis grpc stubs to where they are used

#31 maxdebayser closed 1 month ago
4
Error in generating tokens with example_vllm.py

#30 GhaziSyed closed 4 days ago
1
Could we package fmperf more like a Python-library?

#29 tdoublep opened 3 months ago
1
Support benchmarking of vLLM advanced features

#28 jvlunteren closed 3 months ago
0
Support benchmarking of vLLM advanced features

#27 tdoublep closed 3 months ago
1
Add code option to workload generator

#26 tdoublep opened 4 months ago
1
adding dump of combined results

#25 bringlein closed 4 months ago
0
Unit tests for verifying imports

#24 GhaziSyed closed 3 months ago
7
Enabling prometheus support with service monitors

#23 rohanarora opened 4 months ago
0
Enable ITL, TTFT, E2E latency computation using mean rather than median

#22 tdoublep opened 4 months ago
0
Updates to Kind with GPUs setup

#21 rohanarora opened 4 months ago
2
Adding corresponding import for the realistic workload specification

#20 rohanarora closed 4 months ago
1
Add heterogeneous data-driven workload generator

#19 mal-zurich closed 5 months ago
0
Add some notebooks showing how to plot the data etc

#18 tdoublep opened 5 months ago
1
Add linting action (black)

#17 tdoublep closed 4 months ago
0
Rework names of workload specifications.

#16 tdoublep closed 5 months ago
1
Replace local image with quay.io image

#15 tdoublep closed 5 months ago
0
Support for vLLM --enable-chunked-prefill option

#14 jvlunteren closed 5 months ago
0
fmperf crashes when vLLM server is started using --enable-chunked-prefill option

#13 jvlunteren closed 5 months ago
0
Readme fix

#12 GhaziSyed closed 5 months ago
0
Remove transformers dependency

#11 tdoublep opened 5 months ago
0
fix to issue: vllm model caching #2

#10 GhaziSyed closed 5 months ago
0
Adding the fix for issue: New workloadspec classes #1, and updating e…

#9 GhaziSyed closed 5 months ago
0
make energy metrics as input variables

#8 rinana closed 5 months ago
2
Energy metrics as input variables

#7 rinana closed 5 months ago
1
removing artifcatory id

#6 GhaziSyed closed 5 months ago
0
Add support for server deployment on MIG

#5 yuezhu1 opened 5 months ago
7
More Model API support is needed

#4 wangchen615 opened 5 months ago
1
GitHub actions need to be configured for testing code and automatic benchmarking

#3 wangchen615 opened 5 months ago
3
vllm model caching

#2 GhaziSyed closed 5 months ago
0
New workloadspec classes

#1 GhaziSyed closed 5 months ago
0