Feature Request: EAGLE-4 support

Open
#30,925 0 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
25/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Tech stack
cpp

Research direction

Start by reading the speculative decoding and model architecture support in llama.cpp, then trace where GGUF model metadata and tensors are loaded. The issue identifies EAGLE-4 additions and a drafter-only option, but names no files or tests. Done means supporting those additions and validating the proposed drafter behavior; clarify the source-available license before implementation.

Written by the indexing model from the issue text.

Description

enhancement
Prerequisites
  • I am running the latest code. Mention the version if possible as well.
  • I carefully followed the README.md.
  • I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
  • I reviewed the Discussions, and have a new and useful enhancement to share.
Feature Description

EAGLE-4 came out 20.10, there's a dense-draft Qwen3.8-27B-EAGLE4. Came 4 things in arch: fc_norm for 3 layers, norm_output, SWA 4096, low-rank candidate lm_head. There's an option to do only drafter.

Motivation

On A100: 1.93× at Q8_0, 1.57–1.60× к Q4_K_M. low-rank head: same acceptance, but ~20% faster.
At this model native MTP is a bit faster

License is Eagle 4 Source-Available. Please tell me is this ok?

Possible Implementation

GGUF remains eagle3, options through key and tensors.

My PR is expected

Dominant language
C++
Stars
133k
Forks
25.1k
Avg merge
2d 8h
Merged PRs (30d)
495

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from ggml-org/llama.cpp

All issues in ggml-org/llama.cpp

Similar issues

More C++ issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.