Instructions to use huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560")

# Load model directly
from transformers import AutoTokenizer, LlamaForCausalLMEagle3

tokenizer = AutoTokenizer.from_pretrained("huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560")
model = LlamaForCausalLMEagle3.from_pretrained("huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560

SGLang

How to use huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560 with Docker Model Runner:
```
docker model run hf.co/huluhuluu/qwen3-1p7b-eagle3-k2-sw256-mha-sharegpt-epoch-9-step-171560
```

huluhuluu commited on 29 days ago

Commit

6da28d9

verified ·

1 Parent(s): 280e947

Document training parameters

Browse files

Files changed (1) hide show

README.md +10 -8

README.md CHANGED Viewed

@@ -18,9 +18,9 @@ base_model:
   - Qwen/Qwen3-1.7B
 ---
-# Qwen3-1.7B EAGLE3 K2 SW256 MHA ShareGPT
-This repository contains one SpecForge EAGLE3 draft-model checkpoint for use with `Qwen/Qwen3-1.7B`.
 It is a draft model for speculative decoding, not a standalone target language model.
 ## Checkpoint
@@ -31,16 +31,16 @@ It is a draft model for speculative decoding, not a standalone target language m
 | Checkpoint | `epoch_9_step_171560` |
 | Epoch | `9` |
 | Global step | `171560` |
-| Files | `config.json`, `model.safetensors`, `training_state.pt` |
 ## Training Parameters
 | Parameter | Value |
 |---|---|
 | Base model | `Qwen/Qwen3-1.7B` |
-| Method | SpecForge EAGLE3 online training |
-| Dataset | ShareGPT-format SpecForge training set |
-| Learning rate | `1e-4` |
 | Batch size | `1` |
 | Target batch size | `1` |
 | Epochs configured | `10` |
@@ -48,6 +48,7 @@ It is a draft model for speculative decoding, not a standalone target language m
 | Warmup ratio | `0.015` |
 | Max grad norm | `0.5` |
 | TTT length | `5` |
 | Draft sliding window | `256` |
 | Save interval | `5000` |
 | Eval interval | `5000` |
@@ -74,10 +75,11 @@ It is a draft model for speculative decoding, not a standalone target language m
 | Max position embeddings | `40960` |
 | Sliding window | `256` |
 | Max window layers | `2` |
 ## Notes
-- This is the multi-head-attention variant: `num_key_value_heads=16`.
 - The checkpoint is intended to be loaded by SpecForge/EAGLE3-compatible code.
-- `training_state.pt` is included for provenance and training-state inspection.
 - No benchmark claim is made in this card.

   - Qwen/Qwen3-1.7B
 ---
+# Qwen3 1.7B EAGLE3 K2 SW256 MHA ShareGPT
+This repository contains a SpecForge EAGLE3 draft-model checkpoint for use with `Qwen/Qwen3-1.7B`.
 It is a draft model for speculative decoding, not a standalone target language model.
 ## Checkpoint
 | Checkpoint | `epoch_9_step_171560` |
 | Epoch | `9` |
 | Global step | `171560` |
+| Files | `config.json`, `model.safetensors`, `training_state.pt` (when present) |
 ## Training Parameters
 | Parameter | Value |
 |---|---|
 | Base model | `Qwen/Qwen3-1.7B` |
+| Method | `SpecForge EAGLE3 online training` |
+| Training data | `sharegpt_train.jsonl` |
+| Learning rate | `0.0001` |
 | Batch size | `1` |
 | Target batch size | `1` |
 | Epochs configured | `10` |
 | Warmup ratio | `0.015` |
 | Max grad norm | `0.5` |
 | TTT length | `5` |
+| Draft layers | `2` |
 | Draft sliding window | `256` |
 | Save interval | `5000` |
 | Eval interval | `5000` |
 | Max position embeddings | `40960` |
 | Sliding window | `256` |
 | Max window layers | `2` |
+| Future hidden | `Not recorded` |
 ## Notes
+- This is the highest-step local checkpoint available when this repository was published.
 - The checkpoint is intended to be loaded by SpecForge/EAGLE3-compatible code.
+- `training_state.pt` is included when available for provenance and training-state inspection.
 - No benchmark claim is made in this card.