Instructions to use dogeplusplus/duo-laguna-adapter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dogeplusplus/duo-laguna-adapter with Transformers:

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

pipe = pipeline("text-generation", model="dogeplusplus/duo-laguna-adapter", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("dogeplusplus/duo-laguna-adapter", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("dogeplusplus/duo-laguna-adapter", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use dogeplusplus/duo-laguna-adapter with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/dogeplusplus/duo-laguna-adapter

SGLang

How to use dogeplusplus/duo-laguna-adapter 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 "dogeplusplus/duo-laguna-adapter" \
    --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": "dogeplusplus/duo-laguna-adapter",
		"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 "dogeplusplus/duo-laguna-adapter" \
        --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": "dogeplusplus/duo-laguna-adapter",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use dogeplusplus/duo-laguna-adapter with Docker Model Runner:
```
docker model run hf.co/dogeplusplus/duo-laguna-adapter
```

duo-laguna-adapter / config.json

dogeplusplus

Upload DuoAttention Laguna adapter from rosy-energy-5

7182ebc verified 2 months ago

Raw

History Blame Contribute Delete

4.35 kB

	{
	"_name_or_path": "poolside/Laguna-XS.2",
	"architectures": [
	"DuoLagunaForCausalLM"
	],
	"attention_bias": false,
	"attention_dropout": 0.0,
	"auto_map": {
	"AutoModelForCausalLM": "modeling_duo_laguna.DuoLagunaForCausalLM"
	},
	"bos_token_id": 2,
	"chunk_size_feed_forward": 0,
	"dtype": "bfloat16",
	"duo_attention": {
	"architecture": "laguna",
	"base_config_name_or_path": null,
	"base_model_name_or_path": "poolside/Laguna-XS.2",
	"enabled": true,
	"format_version": 1,
	"full_attention_heads_file": "duo_attention/full_attention_heads.pt",
	"full_attention_heads_tsv_file": "duo_attention/full_attention_heads.tsv",
	"patch_mode": "eval",
	"recent_size": 256,
	"sink_size": 64,
	"training_recent_size": 256,
	"training_sink_size": 64
	},
	"eos_token_id": [
	2,
	24
	],
	"gating": true,
	"head_dim": 128,
	"hidden_act": "silu",
	"hidden_size": 2048,
	"id2label": {
	"0": "LABEL_0",
	"1": "LABEL_1"
	},
	"initializer_range": 0.02,
	"intermediate_size": 8192,
	"is_encoder_decoder": false,
	"label2id": {
	"LABEL_0": 0,
	"LABEL_1": 1
	},
	"layer_types": [
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention",
	"full_attention",
	"sliding_attention",
	"sliding_attention",
	"sliding_attention"
	],
	"max_position_embeddings": 262144,
	"mlp_layer_types": [
	"dense",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse",
	"sparse"
	],
	"model_type": "laguna",
	"moe_apply_router_weight_on_input": false,
	"moe_intermediate_size": 512,
	"moe_routed_scaling_factor": 2.5,
	"moe_router_logit_softcapping": 0.0,
	"num_attention_heads": 48,
	"num_attention_heads_per_layer": [
	48,
	64,
	64,
	64,
	48,
	64,
	64,
	64,
	48,
	64,
	64,
	64,
	48,
	64,
	64,
	64,
	48,
	64,
	64,
	64,
	48,
	64,
	64,
	64,
	48,
	64,
	64,
	64,
	48,
	64,
	64,
	64,
	48,
	64,
	64,
	64,
	48,
	64,
	64,
	64
	],
	"num_experts": 256,
	"num_experts_per_tok": 8,
	"num_hidden_layers": 40,
	"num_key_value_heads": 8,
	"output_attentions": false,
	"output_hidden_states": false,
	"output_router_logits": false,
	"pad_token_id": 9,
	"partial_rotary_factor": 0.5,
	"problem_type": null,
	"return_dict": true,
	"rms_norm_eps": 1e-06,
	"rope_parameters": {
	"full_attention": {
	"attention_factor": 1.0,
	"beta_fast": 64.0,
	"beta_slow": 1.0,
	"factor": 64.0,
	"original_max_position_embeddings": 4096,
	"partial_rotary_factor": 0.5,
	"rope_theta": 500000.0,
	"rope_type": "yarn"
	},
	"original_max_position_embeddings": 4096,
	"rope_type": "default",
	"sliding_attention": {
	"partial_rotary_factor": 1.0,
	"rope_theta": 10000.0,
	"rope_type": "default"
	}
	},
	"router_aux_loss_coef": 0.0,
	"shared_expert_intermediate_size": 512,
	"sliding_window": 512,
	"tie_word_embeddings": false,
	"transformers_version": "5.9.0",
	"use_cache": true,
	"vocab_size": 100352
	}