feat(ci): add continuous batching to benchmarks #41916
McPatate
force pushed
from
c86c753d
to
5065a80b
45 days ago
McPatate
force pushed
from
5065a80b
to
74cf074f
44 days ago
McPatate
force pushed
from
74cf074f
to
50a13aa5
43 days ago
McPatate
force pushed
from
50a13aa5
to
5e25d045
43 days ago
McPatate
force pushed
from
5e25d045
to
cb10e059
43 days ago
remi-or
requested changes
on 2025-11-03
McPatate
force pushed
from
cb10e059
to
c86c753d
40 days ago
McPatate
force pushed
from
c86c753d
to
8ac188bf
40 days ago
McPatate
force pushed
from
442a8c39
to
bfea9d55
40 days ago
McPatate
force pushed
from
80c2076a
to
0ff56066
39 days ago
McPatate
marked this pull request as ready for review 39 days ago
McPatate
force pushed
from
0ff56066
to
ad9fcfe8
39 days ago
McPatate
force pushed
from
b0f1435f
to
81311a7c
37 days ago
remi-or
approved these changes
on 2025-11-07
feat(ci): add continuous batching to benchmarks
ed280fa5
refactor(ci): PR comments
a69f1a87
refactor(cb): when stopping, block by default
b7b10bb8
fix(benchmarks): `stream` -> `streaming`
d5fbaf3e
fix(benchmarks): invalid configuration when cb has attn_impl == sdpa
8b376134
tests(cb): fix attn impl
0848f9e7
fix(benchmarks): update `get_throughput` formula
de76311b
fix(benchmarks): prevent version conflicts and ensure proper cleanup …
bd7bbf8c
fix(benchmarks): raise the exception on failure instead of ignoring
1b1bdc6f
test(cb): comment out failing tests for now
8dd6063c
fix(benchmarks): revert `finally` removal but keep raising exception
5d3f62c4
test(cb): fix missing `require_read_token` import
24d01637
refactor(benchmarks): error if no benchmarks were run
23c9735c
refactor(benchmarks): change default lvls of cb bench config
df9d2a22
McPatate
force pushed
from
72b2fb13
to
df9d2a22
36 days ago
McPatate
enabled auto-merge (squash) 36 days ago
fix(ci): improper git sha checkout out
2d9ca276
McPatate
merged
069684ef
into main 36 days ago
McPatate
deleted the feat/add_cb_to_bench_ci branch 36 days ago
McPatate
restored the head branch 36 days ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub