Skip to content

llama: refactor fused ops#24646

Merged
am17an merged 4 commits into
ggml-org:masterfrom
am17an:refactor-fused-ops
Jul 8, 2026
Merged

llama: refactor fused ops#24646
am17an merged 4 commits into
ggml-org:masterfrom
am17an:refactor-fused-ops

Conversation

@am17an

@am17an am17an commented Jun 15, 2026

Copy link
Copy Markdown
Contributor

Overview

Refactor op detection to be more streamlined, add llm_graph_fused_node so that we don't rely on tensor names. The intention is to make adding new attention ops (e.g. #24231 + various other new interesting ops in dsv4) easier.

Additional information

Requirements

@ggerganov ggerganov left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The LLAMA_TENSOR_NAME_s from llama-impl.h are no longer needed - can be removed

Comment thread src/llama-context.cpp Outdated
Comment thread src/llama-graph.h Outdated
Comment thread src/llama-graph.h Outdated
@am17an

am17an commented Jul 8, 2026

Copy link
Copy Markdown
Contributor Author

@ggml-org/maintainers can I get another approval?

@am17an am17an merged commit 90e0f5c into ggml-org:master Jul 8, 2026
23 of 25 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

model Model specific

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants