Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
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
2 changes: 1 addition & 1 deletion ARCHITECTURE.md
Original file line number Diff line number Diff line change
Expand Up @@ -60,7 +60,7 @@ chat-template tokenization (PR#385).

Full multimodal VLMs: a vision tower (image → patch features) plus a causal text
decoder (features + prompt → generated text). Both towers compile and run on
Spyre; the projector / patch-embed / feature-merge ops that don't lower run on
Spyre, the projector / patch-embed / feature-merge ops that don't lower run on
CPU (see Multimodal VLM Path below).

| Model | model\_type | Towers | Stick Aligned | CPU Accurate | Spyre Compiles | Spyre Runs |
Expand Down
2 changes: 1 addition & 1 deletion tests/model_registry.py
Original file line number Diff line number Diff line change
Expand Up @@ -793,7 +793,7 @@ def _all_paths(
"""
return [
info["path"]
for info in models.values()
for _, info in models.items()
if (include_gated or not info.get("is_gated", False))
and (predicate is None or predicate(info))
]
Expand Down
Loading