Skip to content

Set NVFP4 MSE preset effective bits to 4.5 #2223

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
realAsma wants to merge 1 commit into
base: main
Choose a base branch
from
+5 −0
Open

Set NVFP4 MSE preset effective bits to 4.5 #2223

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions modelopt_recipes/configs/ptq/presets/model/nvfp4_w4a4_weight_mse_fp8_sweep.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,7 @@ imports:
algorithm:
method: mse
fp8_scale_sweep: true
effective_bits: 4.5
quant_cfg:
- $import: base_disable_all
- quantizer_name: '*weight_quantizer'
Expand Down
4 changes: 4 additions & 0 deletions tests/unit/recipe/test_presets.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -63,6 +63,10 @@ def test_kv_none_sentinel_is_not_a_discovered_preset():
assert presets.KV_CACHE_NONE not in presets.KV_QUANT_CFG_CHOICES


def test_nvfp4_mse_preset_uses_nvfp4_storage_cost():
assert presets.QUANT_CFG_CHOICES["nvfp4_w4a4_weight_mse_fp8_sweep"]["effective_bits"] == 4.5


def test_w4a16_nvfp4_preset_disables_vllm_marlin_incompatible_projections():
disabled_quantizers = {
entry["quantizer_name"]
Expand Down
Loading