--- title: Supported Models metatags: description: "SpecBundle supported EAGLE3 draft models for speculative decoding - Llama, Qwen, DeepSeek, GLM, and more." --- ## [Released Models](https://huggingface.co/collections/lmsys/specbundle) We list the models released by the SpecForge and several industrial partners below. These models are released as part of the SpecBundle models, which are trained on large-scale multi-domain datasets and deliver exceptional performance on various benchmarks. > We also include some of the models previously trained by the SpecForge team but not technically part of the SpecBundle release. > We mark models trained on ShareGPT+Ultrachat datasets with a **\*** mark and models trained on Perfect-Blend datasets but released before SpecBundle with **+** mark. ### Llama Series
Target Model EAGLE3 Draft Model
meta-llama/Llama-3.1-8B-Instruct [🤗 Hugging Face](https://huggingface.co/lmsys/SGLang-EAGLE3-Llama-3.1-8B-Instruct-SpecForge)
meta-llama/Llama-3.3-70B-Instruct [🤗 Hugging Face](https://huggingface.co/lmsys/SGLang-EAGLE3-Llama-3.3-70B-Instruct-SpecForge)
meta-llama/Llama-4-Scout-17B-16E-Instruct [🤗 Hugging Face](https://huggingface.co/lmsys/SGLang-EAGLE3-Llama-4-Scout-17B-16E-Instruct-SpecForge)
meta-llama/Llama-4-Maverick-17B-128E-Instruct [🤗 Hugging Face \*](https://huggingface.co/lmsys/sglang-EAGLE3-Llama-4-Maverick-17B-128E-Instruct-v1)
### Qwen Series
Target Model EAGLE3 Draft Model
Qwen/Qwen3-30B-A3B-Instruct-2507 [🤗 Hugging Face](https://huggingface.co/lmsys/SGLang-EAGLE3-Qwen3-30B-A3B-Instruct-2507-SpecForge-Nex)
Qwen/Qwen3-235B-A22B-Instruct-2507 [🤗 Hugging Face](https://huggingface.co/lmsys/SGLang-EAGLE3-Qwen3-235B-A22B-Instruct-2507-SpecForge-Meituan)
Qwen/Qwen3-Next-80B-A3B-Instruct-FP8 [🤗 Hugging Face](https://huggingface.co/lmsys/SGLang-EAGLE3-Qwen3-Next-80B-A3B-Instruct-FP8-perfect-blend-regenerated)
### Qwen Coder Series
Target Model EAGLE3 Draft Model
Qwen/Qwen3-Coder-30B-A3B-Instruct [🤗 Hugging Face](https://huggingface.co/lmsys/SGLang-EAGLE3-Qwen3-Coder-30B-A3B-Instruct-SpecForge)
Qwen/Qwen3-Coder-480B-A35B-Instruct [🤗 Hugging Face](https://huggingface.co/lmsys/SGLang-EAGLE3-Qwen3-Coder-480B-A35B-Instruct-SpecForge-EigenAI)
### Ling Series
Target Model EAGLE3 Draft Model
inclusionAI/Ling-flash-2.0 [🤗 Hugging Face](https://huggingface.co/AQ-MedAI/Ling-Flash-2.0-eagle3)
### Kimi Series
Target Model EAGLE3 Draft Model
moonshotai/Kimi-K2-Instruct [🤗 Hugging Face](https://huggingface.co/AQ-MedAI/Kimi-K2-Instruct-eagle3)
### GPT-OSS Series
Target Model EAGLE3 Draft Model
openai/gpt-oss-20b [🤗 Hugging Face +](https://huggingface.co/zhuyksir/EAGLE3-gpt-oss-20b-bf16)
openai/gpt-oss-120b [🤗 Hugging Face +](https://huggingface.co/lmsys/EAGLE3-gpt-oss-120b-bf16)
### Nex Series
Target Model EAGLE3 Draft Model
nex-agi/Qwen3-30B-A3B-Nex-N1 [🤗 Hugging Face](https://huggingface.co/nex-agi/SGLANG-EAGLE3-Qwen3-30B-A3B-Nex-N1)
nex-agi/Qwen3-32B-Nex-N1 [🤗 Hugging Face](https://huggingface.co/nex-agi/SGLANG-EAGLE3-Qwen3-32B-Nex-N1)