Instructions to use realYinkaIyiola/Deepseek-R1-Distill-14B-Code with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use realYinkaIyiola/Deepseek-R1-Distill-14B-Code with Transformers:

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

pipe = pipeline("text-generation", model="realYinkaIyiola/Deepseek-R1-Distill-14B-Code")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("realYinkaIyiola/Deepseek-R1-Distill-14B-Code")
model = AutoModelForCausalLM.from_pretrained("realYinkaIyiola/Deepseek-R1-Distill-14B-Code", device_map="auto")
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 realYinkaIyiola/Deepseek-R1-Distill-14B-Code with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "realYinkaIyiola/Deepseek-R1-Distill-14B-Code"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "realYinkaIyiola/Deepseek-R1-Distill-14B-Code",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/realYinkaIyiola/Deepseek-R1-Distill-14B-Code

SGLang

How to use realYinkaIyiola/Deepseek-R1-Distill-14B-Code 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 "realYinkaIyiola/Deepseek-R1-Distill-14B-Code" \
    --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": "realYinkaIyiola/Deepseek-R1-Distill-14B-Code",
		"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 "realYinkaIyiola/Deepseek-R1-Distill-14B-Code" \
        --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": "realYinkaIyiola/Deepseek-R1-Distill-14B-Code",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use realYinkaIyiola/Deepseek-R1-Distill-14B-Code with Docker Model Runner:
```
docker model run hf.co/realYinkaIyiola/Deepseek-R1-Distill-14B-Code
```

Deepseek-R1-Distill-14B-Code / trainer_state.json

realYinkaIyiola

Upload folder using huggingface_hub

ca78b15 verified over 1 year ago

Raw

History Blame Contribute Delete

7.35 kB

	{
	"best_global_step": null,
	"best_metric": null,
	"best_model_checkpoint": null,
	"epoch": 3.0,
	"eval_steps": 1000,
	"global_step": 3360,
	"is_hyper_param_search": false,
	"is_local_process_zero": true,
	"is_world_process_zero": true,
	"log_history": [
	{
	"epoch": 0.08928571428571429,
	"grad_norm": 0.42009279659859,
	"learning_rate": 2.9761904761904763e-06,
	"loss": 0.5182,
	"step": 100
	},
	{
	"epoch": 0.17857142857142858,
	"grad_norm": 0.38633756820312576,
	"learning_rate": 5.9523809523809525e-06,
	"loss": 0.4851,
	"step": 200
	},
	{
	"epoch": 0.26785714285714285,
	"grad_norm": 0.45962638823829793,
	"learning_rate": 8.92857142857143e-06,
	"loss": 0.4816,
	"step": 300
	},
	{
	"epoch": 0.35714285714285715,
	"grad_norm": 0.4226066822293375,
	"learning_rate": 9.988952191691925e-06,
	"loss": 0.4868,
	"step": 400
	},
	{
	"epoch": 0.44642857142857145,
	"grad_norm": 0.3827199419252431,
	"learning_rate": 9.927604254015586e-06,
	"loss": 0.4895,
	"step": 500
	},
	{
	"epoch": 0.5357142857142857,
	"grad_norm": 0.446330637822688,
	"learning_rate": 9.81312123475006e-06,
	"loss": 0.49,
	"step": 600
	},
	{
	"epoch": 0.625,
	"grad_norm": 0.38997662785153187,
	"learning_rate": 9.646737621134112e-06,
	"loss": 0.4875,
	"step": 700
	},
	{
	"epoch": 0.7142857142857143,
	"grad_norm": 0.3660382818183833,
	"learning_rate": 9.430247552150673e-06,
	"loss": 0.4856,
	"step": 800
	},
	{
	"epoch": 0.8035714285714286,
	"grad_norm": 0.3644345773145445,
	"learning_rate": 9.165985472062245e-06,
	"loss": 0.4846,
	"step": 900
	},
	{
	"epoch": 0.8928571428571429,
	"grad_norm": 0.3563534770415374,
	"learning_rate": 8.856800957755e-06,
	"loss": 0.4844,
	"step": 1000
	},
	{
	"epoch": 0.8928571428571429,
	"eval_loss": 0.4723494052886963,
	"eval_runtime": 527.1023,
	"eval_samples_per_second": 3.777,
	"eval_steps_per_second": 0.472,
	"step": 1000
	},
	{
	"epoch": 0.9821428571428571,
	"grad_norm": 0.3915169737374782,
	"learning_rate": 8.50602799133199e-06,
	"loss": 0.4835,
	"step": 1100
	},
	{
	"epoch": 1.0714285714285714,
	"grad_norm": 0.3986100916298589,
	"learning_rate": 8.117449009293668e-06,
	"loss": 0.4388,
	"step": 1200
	},
	{
	"epoch": 1.1607142857142858,
	"grad_norm": 0.38571453527776084,
	"learning_rate": 7.69525411596865e-06,
	"loss": 0.4311,
	"step": 1300
	},
	{
	"epoch": 1.25,
	"grad_norm": 0.3530259346548572,
	"learning_rate": 7.243995901002312e-06,
	"loss": 0.427,
	"step": 1400
	},
	{
	"epoch": 1.3392857142857144,
	"grad_norm": 0.4069536036588859,
	"learning_rate": 6.768540348112908e-06,
	"loss": 0.429,
	"step": 1500
	},
	{
	"epoch": 1.4285714285714286,
	"grad_norm": 0.40040431306856655,
	"learning_rate": 6.274014364473274e-06,
	"loss": 0.4282,
	"step": 1600
	},
	{
	"epoch": 1.5178571428571428,
	"grad_norm": 0.34105714911676593,
	"learning_rate": 5.765750496516547e-06,
	"loss": 0.4318,
	"step": 1700
	},
	{
	"epoch": 1.6071428571428572,
	"grad_norm": 0.3679992579349683,
	"learning_rate": 5.249229428303486e-06,
	"loss": 0.4272,
	"step": 1800
	},
	{
	"epoch": 1.6964285714285714,
	"grad_norm": 0.37091945305243507,
	"learning_rate": 4.730020882499964e-06,
	"loss": 0.4292,
	"step": 1900
	},
	{
	"epoch": 1.7857142857142856,
	"grad_norm": 0.4210486903676345,
	"learning_rate": 4.213723561238074e-06,
	"loss": 0.4258,
	"step": 2000
	},
	{
	"epoch": 1.7857142857142856,
	"eval_loss": 0.472212553024292,
	"eval_runtime": 526.9195,
	"eval_samples_per_second": 3.779,
	"eval_steps_per_second": 0.473,
	"step": 2000
	},
	{
	"epoch": 1.875,
	"grad_norm": 0.4028542180886112,
	"learning_rate": 3.705904774487396e-06,
	"loss": 0.4307,
	"step": 2100
	},
	{
	"epoch": 1.9642857142857144,
	"grad_norm": 0.4621878586501629,
	"learning_rate": 3.2120404069325695e-06,
	"loss": 0.4271,
	"step": 2200
	},
	{
	"epoch": 2.0535714285714284,
	"grad_norm": 0.4678503190324282,
	"learning_rate": 2.737455870703155e-06,
	"loss": 0.3889,
	"step": 2300
	},
	{
	"epoch": 2.142857142857143,
	"grad_norm": 0.4575407722156448,
	"learning_rate": 2.2872686806712037e-06,
	"loss": 0.374,
	"step": 2400
	},
	{
	"epoch": 2.232142857142857,
	"grad_norm": 0.4241549942755108,
	"learning_rate": 1.8663332715355399e-06,
	"loss": 0.3694,
	"step": 2500
	},
	{
	"epoch": 2.3214285714285716,
	"grad_norm": 0.4261640971882884,
	"learning_rate": 1.4791886517382415e-06,
	"loss": 0.3695,
	"step": 2600
	},
	{
	"epoch": 2.4107142857142856,
	"grad_norm": 0.4640732620056673,
	"learning_rate": 1.1300094586688632e-06,
	"loss": 0.3719,
	"step": 2700
	},
	{
	"epoch": 2.5,
	"grad_norm": 0.3661173371298237,
	"learning_rate": 8.225609429353187e-07,
	"loss": 0.3749,
	"step": 2800
	},
	{
	"epoch": 2.5892857142857144,
	"grad_norm": 0.4357767645022839,
	"learning_rate": 5.601583671126532e-07,
	"loss": 0.3702,
	"step": 2900
	},
	{
	"epoch": 2.678571428571429,
	"grad_norm": 0.41225890414805616,
	"learning_rate": 3.4563125677897936e-07,
	"loss": 0.3709,
	"step": 3000
	},
	{
	"epoch": 2.678571428571429,
	"eval_loss": 0.48118627071380615,
	"eval_runtime": 526.9164,
	"eval_samples_per_second": 3.779,
	"eval_steps_per_second": 0.473,
	"step": 3000
	},
	{
	"epoch": 2.767857142857143,
	"grad_norm": 0.5194673179252723,
	"learning_rate": 1.8129288932490276e-07,
	"loss": 0.3689,
	"step": 3100
	},
	{
	"epoch": 2.857142857142857,
	"grad_norm": 0.4488783086501226,
	"learning_rate": 6.891534954310886e-08,
	"loss": 0.3666,
	"step": 3200
	},
	{
	"epoch": 2.946428571428571,
	"grad_norm": 0.3580886076427873,
	"learning_rate": 9.710420977340763e-09,
	"loss": 0.3702,
	"step": 3300
	},
	{
	"epoch": 3.0,
	"step": 3360,
	"total_flos": 3000194903834624.0,
	"train_loss": 0.4290710630871001,
	"train_runtime": 89946.115,
	"train_samples_per_second": 0.597,
	"train_steps_per_second": 0.037
	}
	],
	"logging_steps": 100,
	"max_steps": 3360,
	"num_input_tokens_seen": 0,
	"num_train_epochs": 3,
	"save_steps": 1000,
	"stateful_callbacks": {
	"TrainerControl": {
	"args": {
	"should_epoch_stop": false,
	"should_evaluate": false,
	"should_log": false,
	"should_save": true,
	"should_training_stop": true
	},
	"attributes": {}
	}
	},
	"total_flos": 3000194903834624.0,
	"train_batch_size": 2,
	"trial_name": null,
	"trial_params": null
	}