Instructions to use rqadri/deepmath4b_100s with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use rqadri/deepmath4b_100s with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("unsloth/Qwen3-4B-Instruct-2507")
model = PeftModel.from_pretrained(base_model, "rqadri/deepmath4b_100s")

Transformers

How to use rqadri/deepmath4b_100s with Transformers:

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

pipe = pipeline("text-generation", model="rqadri/deepmath4b_100s")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("rqadri/deepmath4b_100s", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use rqadri/deepmath4b_100s with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rqadri/deepmath4b_100s"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rqadri/deepmath4b_100s",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/rqadri/deepmath4b_100s

SGLang

How to use rqadri/deepmath4b_100s 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 "rqadri/deepmath4b_100s" \
    --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": "rqadri/deepmath4b_100s",
		"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 "rqadri/deepmath4b_100s" \
        --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": "rqadri/deepmath4b_100s",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use rqadri/deepmath4b_100s with Docker Model Runner:
```
docker model run hf.co/rqadri/deepmath4b_100s
```

deepmath4b_100s / trainer_state.json

rqadri

Upload folder using huggingface_hub

e86739c verified about 1 year ago

Raw History Blame Contribute Delete

2.33 kB

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0019414459889725867,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 512.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 512.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 512.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 0.0009707229944862934,
      "frac_reward_zero_std": 0.61,
      "grad_norm": 0.02464698627591133,
      "kl": 0.03792417001328431,
      "learning_rate": 4.9000000000000005e-06,
      "loss": 0.0,
      "num_tokens": 222000.0,
      "reward": 0.43,
      "reward_std": 0.20350852847099304,
      "rewards/compute_reward/mean": 0.43,
      "rewards/compute_reward/std": 0.3326409870386124,
      "step": 50
    },
    {
      "completion_length": 512.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 512.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 512.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 512.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 0.0019414459889725867,
      "frac_reward_zero_std": 0.67,
      "grad_norm": 0.020398980006575584,
      "kl": 0.27587527528405187,
      "learning_rate": 4.855143631968242e-06,
      "loss": 0.0003,
      "num_tokens": 444000.0,
      "reward": 0.4275,
      "reward_std": 0.17582903623580934,
      "rewards/compute_reward/mean": 0.4275,
      "rewards/compute_reward/std": 0.40157180368900297,
      "step": 100
    }
  ],
  "logging_steps": 50,
  "max_steps": 500,
  "num_input_tokens_seen": 444000,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}

	{
	"best_global_step": null,
	"best_metric": null,
	"best_model_checkpoint": null,
	"epoch": 0.0019414459889725867,
	"eval_steps": 500,
	"global_step": 100,
	"is_hyper_param_search": false,
	"is_local_process_zero": true,
	"is_world_process_zero": true,
	"log_history": [
	{
	"completion_length": 512.0,
	"completions/clipped_ratio": 1.0,
	"completions/max_length": 512.0,
	"completions/max_terminated_length": 0.0,
	"completions/mean_length": 512.0,
	"completions/mean_terminated_length": 0.0,
	"completions/min_length": 512.0,
	"completions/min_terminated_length": 0.0,
	"epoch": 0.0009707229944862934,
	"frac_reward_zero_std": 0.61,
	"grad_norm": 0.02464698627591133,
	"kl": 0.03792417001328431,
	"learning_rate": 4.9000000000000005e-06,
	"loss": 0.0,
	"num_tokens": 222000.0,
	"reward": 0.43,
	"reward_std": 0.20350852847099304,
	"rewards/compute_reward/mean": 0.43,
	"rewards/compute_reward/std": 0.3326409870386124,
	"step": 50
	},
	{
	"completion_length": 512.0,
	"completions/clipped_ratio": 1.0,
	"completions/max_length": 512.0,
	"completions/max_terminated_length": 0.0,
	"completions/mean_length": 512.0,
	"completions/mean_terminated_length": 0.0,
	"completions/min_length": 512.0,
	"completions/min_terminated_length": 0.0,
	"epoch": 0.0019414459889725867,
	"frac_reward_zero_std": 0.67,
	"grad_norm": 0.020398980006575584,
	"kl": 0.27587527528405187,
	"learning_rate": 4.855143631968242e-06,
	"loss": 0.0003,
	"num_tokens": 444000.0,
	"reward": 0.4275,
	"reward_std": 0.17582903623580934,
	"rewards/compute_reward/mean": 0.4275,
	"rewards/compute_reward/std": 0.40157180368900297,
	"step": 100
	}
	],
	"logging_steps": 50,
	"max_steps": 500,
	"num_input_tokens_seen": 444000,
	"num_train_epochs": 1,
	"save_steps": 50,
	"stateful_callbacks": {
	"TrainerControl": {
	"args": {
	"should_epoch_stop": false,
	"should_evaluate": false,
	"should_log": false,
	"should_save": true,
	"should_training_stop": false
	},
	"attributes": {}
	}
	},
	"total_flos": 0.0,
	"train_batch_size": 4,
	"trial_name": null,
	"trial_params": null
	}