ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Studio diffusion (Phase 11) review round 2: keep GH200/B300 on the fp8 ladder

Browse source 
Codex review: _DATACENTER_GPU_TOKENS omitted GH200 (Grace-Hopper) and B300
(Blackwell Ultra), though it has the distinct GB200/GB300 superchip tokens. So
_is_consumer_gpu returned True for 'NVIDIA GH200 480GB' / 'NVIDIA B300', and the
auto ladder moved int8 ahead of fp8 on those data-center parts -- contradicting
llama_cpp.py's datacenter regex, which lists both. Added GH200 and B300 so they
are treated as data-center class and keep the intended fp8-first behavior.

Test: extends the datacenter parametrize with 'NVIDIA B300' and
'NVIDIA GH200 480GB' (now _is_consumer_gpu False).
This commit is contained in:
Daniel Han 2026-06-29 10:44:20 +00:00
commit 62678e58a8
2 changed files with 5 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

4
studio/backend/core/inference/diffusion_transformer_quant.py
View file

@ -77,13 +77,15 @@ _DATACENTER_GPU_TOKENS = frozenset(
{
"B200",
"B100",
"B300", # Blackwell Ultra data center
"GB200",
"GB300",
"GB10", # Blackwell data center
"H200",
"H100",
"H800",
"H20", # Hopper data center
"H20",
"GH200", # Grace-Hopper superchip (data center)
"A100",
"A800",
"A30",

2
studio/backend/tests/test_diffusion_transformer_quant.py
View file

@ -279,6 +279,8 @@ def test_is_consumer_gpu_true(monkeypatch, name):
"name",
[
"NVIDIA B200",
"NVIDIA B300", # Blackwell Ultra (matches llama_cpp datacenter regex)
"NVIDIA GH200 480GB", # Grace-Hopper superchip (was misread as consumer)
"NVIDIA H100 80GB HBM3",
"NVIDIA A100-SXM4-80GB",
"NVIDIA A40", # data-center Ampere (distinct token from RTX A4000)