Instructions to use peerrh/treeflash-qwen3-coder-30b-a3b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use peerrh/treeflash-qwen3-coder-30b-a3b with Transformers:

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

pipe = pipeline("text-generation", model="peerrh/treeflash-qwen3-coder-30b-a3b", trust_remote_code=True, device_map="auto")

# Load model directly
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("peerrh/treeflash-qwen3-coder-30b-a3b", trust_remote_code=True)
model = AutoModel.from_pretrained("peerrh/treeflash-qwen3-coder-30b-a3b", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use peerrh/treeflash-qwen3-coder-30b-a3b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "peerrh/treeflash-qwen3-coder-30b-a3b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "peerrh/treeflash-qwen3-coder-30b-a3b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/peerrh/treeflash-qwen3-coder-30b-a3b

SGLang

How to use peerrh/treeflash-qwen3-coder-30b-a3b 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 "peerrh/treeflash-qwen3-coder-30b-a3b" \
    --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": "peerrh/treeflash-qwen3-coder-30b-a3b",
		"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 "peerrh/treeflash-qwen3-coder-30b-a3b" \
        --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": "peerrh/treeflash-qwen3-coder-30b-a3b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use peerrh/treeflash-qwen3-coder-30b-a3b with Docker Model Runner:
```
docker model run hf.co/peerrh/treeflash-qwen3-coder-30b-a3b
```

treeflash-qwen3-coder-30b-a3b / config.json

peerrh

Add files using upload-large-folder tool

74502d9 verified about 1 month ago

Raw

History Blame Contribute Delete

1.87 kB

	{
	"_name_or_path": "z-lab/Qwen3-Coder-30B-A3B-DFlash",
	"ar_approximation": "swiglu",
	"architectures": [
	"TreeFlashDraftModel"
	],
	"attention_bias": false,
	"attention_dropout": 0.0,
	"auto_map": {
	"AutoModel": "tree_flash.TreeFlashDraftModel"
	},
	"block_size": 16,
	"bos_token_id": null,
	"candidate_tokens": 16,
	"chunk_size_feed_forward": 0,
	"dflash_config": {
	"mask_token_id": 151669,
	"target_layer_ids": [
	1,
	12,
	23,
	34,
	45
	]
	},
	"dtype": "bfloat16",
	"eos_token_id": 151645,
	"head_dim": 128,
	"hidden_act": "silu",
	"hidden_size": 2048,
	"id2label": {
	"0": "LABEL_0",
	"1": "LABEL_1"
	},
	"initializer_range": 0.02,
	"intermediate_size": 6144,
	"is_encoder_decoder": false,
	"label2id": {
	"LABEL_0": 0,
	"LABEL_1": 1
	},
	"layer_types": [
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention",
	"full_attention"
	],
	"max_position_embeddings": 262144,
	"max_window_layers": 8,
	"model_type": "qwen3",
	"num_attention_heads": 32,
	"num_hidden_layers": 8,
	"num_key_value_heads": 4,
	"num_target_layers": 48,
	"output_attentions": false,
	"output_hidden_states": false,
	"pad_token_id": 151643,
	"problem_type": null,
	"return_dict": true,
	"rms_norm_eps": 1e-06,
	"rope_parameters": {
	"rope_theta": 10000000,
	"rope_type": "default"
	},
	"sliding_window": null,
	"tie_word_embeddings": false,
	"transformers_version": "5.8.1",
	"treeflash_config": {
	"ar_approximation": "swiglu",
	"base_model_name": "Qwen/Qwen3-Coder-30B-A3B-Instruct",
	"candidate_tokens": 16,
	"dflash_checkpoint": "z-lab/Qwen3-Coder-30B-A3B-DFlash",
	"top_m": null
	},
	"use_cache": true,
	"use_sliding_window": false,
	"vocab_size": 151936
	}