Instructions to use k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8 with Transformers:

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

pipe = pipeline("text-generation", model="k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8")
model = AutoModelForMultimodalLM.from_pretrained("k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8

SGLang

How to use k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8 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 "k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8 with Docker Model Runner:
```
docker model run hf.co/k1h0/DeepSeek-Coder-7B-Instruct-query_nsx_under8
```

DeepSeek-Coder-7B-Instruct-query_nsx_under8 / llamaboard_config.yaml

k1h0

Upload folder using huggingface_hub

29e2abb verified 12 months ago

Raw

History Blame Contribute Delete

2 kB

	top.booster: liger_kernel
	top.checkpoint_path: null
	top.finetuning_type: freeze
	top.model_name: DeepSeek-Coder-7B-Instruct
	top.quantization_bit: none
	top.quantization_method: bitsandbytes
	top.rope_scaling: llama3
	top.template: deepseekcoder
	train.additional_target: ''
	train.apollo_rank: 256
	train.apollo_scale: 1
	train.apollo_target: all
	train.apollo_update_interval: 200
	train.badam_mode: layer
	train.badam_switch_interval: 50
	train.badam_switch_mode: ascending
	train.badam_update_ratio: 0.05
	train.batch_size: 16
	train.compute_type: bf16
	train.create_new_adapter: false
	train.cutoff_len: 4096
	train.dataset:
	- codes_nsx_under8
	train.dataset_dir: data
	train.ds_offload: false
	train.ds_stage: none
	train.extra_args: '{}'
	train.freeze_extra_modules: ''
	train.freeze_trainable_layers: 2
	train.freeze_trainable_modules: all
	train.galore_rank: 16
	train.galore_scale: 2
	train.galore_target: all
	train.galore_update_interval: 200
	train.gradient_accumulation_steps: 8
	train.learning_rate: 5e-5
	train.logging_steps: 1
	train.lora_alpha: 16
	train.lora_dropout: 0
	train.lora_rank: 8
	train.lora_target: ''
	train.loraplus_lr_ratio: 0
	train.lr_scheduler_type: cosine
	train.mask_history: false
	train.max_grad_norm: '1.0'
	train.max_samples: '50000000'
	train.neat_packing: true
	train.neftune_alpha: 0
	train.num_train_epochs: '1'
	train.packing: true
	train.ppo_score_norm: false
	train.ppo_whiten_rewards: false
	train.pref_beta: 0.1
	train.pref_ftx: 0
	train.pref_loss: sigmoid
	train.report_to:
	- none
	train.resize_vocab: false
	train.reward_model: null
	train.save_steps: 1000
	train.swanlab_api_key: ''
	train.swanlab_mode: cloud
	train.swanlab_project: llamafactory
	train.swanlab_run_name: ''
	train.swanlab_workspace: ''
	train.train_on_prompt: false
	train.training_stage: Supervised Fine-Tuning
	train.use_apollo: true
	train.use_badam: false
	train.use_dora: false
	train.use_galore: false
	train.use_llama_pro: true
	train.use_pissa: false
	train.use_rslora: false
	train.use_swanlab: false
	train.val_size: 0
	train.warmup_steps: 0