eagle3 开源项目推荐（13 个）

sgl-project /SpecForge

轻松训练推测解码模型，并顺利移植到SGLang服务。

lightseekorg /TorchSpec

A PyTorch native library for training speculative decoding models

inclusionAI /ASpec

A high-throughput training framework for Speculative decoding models.

carlosfundora /sglang-1-bit-turbo

AMD ROCm (gfx1030) inference fork with RotorQuant/TurboQuant KV compression, PHANTOM-X zero-copy draft speculation, EAGLE3 speculative decoding, 12 RDNA2 crash fixes, and PrismML Bonsai Q1_0_G128 1-bit GGUF support.

SamsungLabs /WhiFlash

This is a research project to efficiently accelerate speculative decoding with token-level cross-paradigm routing, accepted at EMNLP 2026.

JayYun98 /Draftfit

Train and adapt speculative decoding drafts for LLM inference: DSpark, DFlash, DFlash2, EAGLE3, and official EAGLE4 checkpoints.

tails-mpt /SpecJAX

JAX/TPU-native training framework for EAGLE3 speculative decoding draft models

patrickbdevaney /qwen3-vl-tensorrt-edge-llm

Deployment guide: Qwen3-VL-8B on Jetson AGX Thor with NVIDIA TensorRT-Edge-LLM — NVFP4 quantization and ONNX export on x86, native TensorRT engine build on Thor, EAGLE-3 speculative decoding (67–92 tok/s)

shreyansh26 /qwen-spec-decode-benchmarking

Some benchmarks on running speculative decoding with Qwen-3.6-35B-A3B with EAGLE3, DFlash and MTP.

rohseh303 /rust-eagle3

Controlled EAGLE-3 draft adaptation for frozen Qwen3-8B: Rust latency, replication, scaling controls, and serving trade-offs.

TiaoFeng /spark-eagle3-forge

Retrained EAGLE-3 draft head for Spark-X2.6-4B: draft acceptance on Chinese prompts. Rebuilt 48K draft vocabulary, trained on the deployed Q6_K weights.

patrickbdevaney /mistral-inference

Mistral Small 4 119B NVFP4 + EAGLE-3 on NVIDIA Jetson AGX Thor with vLLM — TRITON_MLA, chat/reasoning_effort patches for the Jetson image, 33.7 tok/s (1.93x)

dmagog /eagle3-speculative-decoding

Воспроизведение EAGLE-3 на Qwen3: где спекулятивное декодирование ускоряет, а где мешает. Два ноутбука, полный прогон на Kaggle T4.