Instructions to use nazdef/1gpu-llm-small-en-it-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nazdef/1gpu-llm-small-en-it-base with Transformers:

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

pipe = pipeline("text-generation", model="nazdef/1gpu-llm-small-en-it-base")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nazdef/1gpu-llm-small-en-it-base")
model = AutoModelForCausalLM.from_pretrained("nazdef/1gpu-llm-small-en-it-base")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use nazdef/1gpu-llm-small-en-it-base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nazdef/1gpu-llm-small-en-it-base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nazdef/1gpu-llm-small-en-it-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/nazdef/1gpu-llm-small-en-it-base

SGLang

How to use nazdef/1gpu-llm-small-en-it-base 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 "nazdef/1gpu-llm-small-en-it-base" \
    --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": "nazdef/1gpu-llm-small-en-it-base",
		"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 "nazdef/1gpu-llm-small-en-it-base" \
        --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": "nazdef/1gpu-llm-small-en-it-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use nazdef/1gpu-llm-small-en-it-base with Docker Model Runner:
```
docker model run hf.co/nazdef/1gpu-llm-small-en-it-base
```

1gpu-llm-small-en-it-base / decoding_summary.json

nazdef

Add files using upload-large-folder tool

0982b30 verified 15 days ago

Raw

History Blame Contribute Delete

12.9 kB

	{
	"name": "decoding_tuning_en_it_v1",
	"config_path": "/tmp/llm-nanochat-dev-p0p1/configs/eval/decoding_tuning.yaml",
	"checkpoint_path": "/mnt/apps/llm-nanochat/checkpoints/20260622_resume-gpt2small-gpt2preln-k20-wsds800-final2e5-webwiki-step8000-dense50/step_8600.pt",
	"tokenizer_dir": "/mnt/apps/llm-nanochat/tokenizers/tokenizer_20260515_en50it50_webwiki_stratified_500M",
	"tuning_prompts_path": "/tmp/llm-nanochat-dev-p0p1/eval_prompts/decoding_tuning.jsonl",
	"holdout_prompts_path": "/tmp/llm-nanochat-dev-p0p1/eval_prompts/decoding_tuning_holdout.jsonl",
	"seeds": [
	1337,
	1338,
	1339
	],
	"holdout_top_k": 3,
	"target_length_ratio": 0.8,
	"score_weights": {
	"prompt_pass_rate": 2.0,
	"completion_rate": 0.5,
	"distinct_2": 1.2,
	"language_consistency": 0.75,
	"length_closeness": 0.5,
	"loop_rate": -2.0,
	"repeated_4gram_rate": -1.5,
	"language_switch_rate": -0.75
	},
	"presets": [
	{
	"name": "balanced",
	"max_new_tokens": 64,
	"temperature": 0.8,
	"top_k": 50,
	"top_p": 0.95,
	"repetition_penalty": 1.1,
	"no_repeat_ngram_size": 0,
	"score": 2.7889629329004326,
	"score_breakdown": {
	"prompt_pass_rate": 0.08333333333333333,
	"completion_rate": 0.9523809523809523,
	"distinct_2": 0.9175757575757576,
	"language_consistency": 0.9513888888888888,
	"length_closeness": 0.9486607142857142,
	"loop_rate": 0.0,
	"repeated_4gram_rate": 0.09523809523809523,
	"language_switch_rate": 0.0
	},
	"total_rows": 21,
	"scored_rows": 12,
	"passed_rows": 1,
	"prompt_pass_rate": 0.08333333333333333,
	"completion_rate": 0.9523809523809523,
	"avg_generated_tokens": 48.57142857142857,
	"length_ratio": 0.7589285714285714,
	"length_closeness": 0.9486607142857142,
	"distinct_1": 0.5171597633136095,
	"distinct_2": 0.9175757575757576,
	"repeated_4gram_rate": 0.09523809523809523,
	"loop_rate": 0.0,
	"language_switch_rate_en": 0.0,
	"language_switch_rate_it": 0.0,
	"language_consistency_en": 0.9583333333333334,
	"language_consistency_it": 0.9444444444444444,
	"language_consistency_mean": 0.9513888888888888,
	"language_switch_rate_mean": 0.0,
	"tuning_rank": 1
	},
	{
	"name": "creative",
	"max_new_tokens": 64,
	"temperature": 1.0,
	"top_k": 100,
	"top_p": 0.95,
	"repetition_penalty": 1.1,
	"no_repeat_ngram_size": 0,
	"score": 2.7750582298136646,
	"score_breakdown": {
	"prompt_pass_rate": 0.08333333333333333,
	"completion_rate": 1.0,
	"distinct_2": 0.9565217391304348,
	"language_consistency": 0.9791666666666667,
	"length_closeness": 0.9285714285714287,
	"loop_rate": 0.047619047619047616,
	"repeated_4gram_rate": 0.09523809523809523,
	"language_switch_rate": 0.0
	},
	"total_rows": 21,
	"scored_rows": 12,
	"passed_rows": 1,
	"prompt_pass_rate": 0.08333333333333333,
	"completion_rate": 1.0,
	"avg_generated_tokens": 54.857142857142854,
	"length_ratio": 0.8571428571428571,
	"length_closeness": 0.9285714285714287,
	"distinct_1": 0.5599128540305011,
	"distinct_2": 0.9565217391304348,
	"repeated_4gram_rate": 0.09523809523809523,
	"loop_rate": 0.047619047619047616,
	"language_switch_rate_en": 0.0,
	"language_switch_rate_it": 0.0,
	"language_consistency_en": 0.9583333333333334,
	"language_consistency_it": 1.0,
	"language_consistency_mean": 0.9791666666666667,
	"language_switch_rate_mean": 0.0,
	"tuning_rank": 2
	},
	{
	"name": "anti_loop",
	"max_new_tokens": 64,
	"temperature": 0.8,
	"top_k": 50,
	"top_p": 0.9,
	"repetition_penalty": 1.15,
	"no_repeat_ngram_size": 4,
	"score": 2.7175428468801313,
	"score_breakdown": {
	"prompt_pass_rate": 0.0,
	"completion_rate": 0.9523809523809523,
	"distinct_2": 0.9261083743842364,
	"language_consistency": 0.9583333333333333,
	"length_closeness": 0.9654017857142857,
	"loop_rate": 0.0,
	"repeated_4gram_rate": 0.047619047619047616,
	"language_switch_rate": 0.0
	},
	"total_rows": 21,
	"scored_rows": 12,
	"passed_rows": 0,
	"prompt_pass_rate": 0.0,
	"completion_rate": 0.9523809523809523,
	"avg_generated_tokens": 49.42857142857143,
	"length_ratio": 0.7723214285714286,
	"length_closeness": 0.9654017857142857,
	"distinct_1": 0.5252403846153846,
	"distinct_2": 0.9261083743842364,
	"repeated_4gram_rate": 0.047619047619047616,
	"loop_rate": 0.0,
	"language_switch_rate_en": 0.0,
	"language_switch_rate_it": 0.0,
	"language_consistency_en": 0.9166666666666666,
	"language_consistency_it": 1.0,
	"language_consistency_mean": 0.9583333333333333,
	"language_switch_rate_mean": 0.0,
	"tuning_rank": 3
	},
	{
	"name": "conservative",
	"max_new_tokens": 64,
	"temperature": 0.6,
	"top_k": 40,
	"top_p": 0.9,
	"repetition_penalty": 1.05,
	"no_repeat_ngram_size": 0,
	"score": 1.3666469369276413,
	"score_breakdown": {
	"prompt_pass_rate": 0.25,
	"completion_rate": 0.8571428571428571,
	"distinct_2": 0.6885441527446301,
	"language_consistency": 0.9353070175438596,
	"length_closeness": 0.9635416666666666,
	"loop_rate": 0.2857142857142857,
	"repeated_4gram_rate": 0.6666666666666666,
	"language_switch_rate": 0.0
	},
	"total_rows": 21,
	"scored_rows": 12,
	"passed_rows": 3,
	"prompt_pass_rate": 0.25,
	"completion_rate": 0.8571428571428571,
	"avg_generated_tokens": 49.333333333333336,
	"length_ratio": 0.7708333333333334,
	"length_closeness": 0.9635416666666666,
	"distinct_1": 0.375,
	"distinct_2": 0.6885441527446301,
	"repeated_4gram_rate": 0.6666666666666666,
	"loop_rate": 0.2857142857142857,
	"language_switch_rate_en": 0.0,
	"language_switch_rate_it": 0.0,
	"language_consistency_en": 0.8706140350877193,
	"language_consistency_it": 1.0,
	"language_consistency_mean": 0.9353070175438596,
	"language_switch_rate_mean": 0.0,
	"tuning_rank": 4
	},
	{
	"name": "greedy",
	"max_new_tokens": 64,
	"temperature": 0.0,
	"top_k": 0,
	"top_p": 1.0,
	"repetition_penalty": 1.0,
	"no_repeat_ngram_size": 0,
	"score": -0.9418192340067341,
	"score_breakdown": {
	"prompt_pass_rate": 0.0,
	"completion_rate": 0.8571428571428571,
	"distinct_2": 0.05499438832772166,
	"language_consistency": 0.7708333333333333,
	"length_closeness": 0.9709821428571428,
	"loop_rate": 0.7142857142857143,
	"repeated_4gram_rate": 0.7142857142857143,
	"language_switch_rate": 0.0
	},
	"total_rows": 21,
	"scored_rows": 12,
	"passed_rows": 0,
	"prompt_pass_rate": 0.0,
	"completion_rate": 0.8571428571428571,
	"avg_generated_tokens": 49.714285714285715,
	"length_ratio": 0.7767857142857143,
	"length_closeness": 0.9709821428571428,
	"distinct_1": 0.0429042904290429,
	"distinct_2": 0.05499438832772166,
	"repeated_4gram_rate": 0.7142857142857143,
	"loop_rate": 0.7142857142857143,
	"language_switch_rate_en": 0.0,
	"language_switch_rate_it": 0.0,
	"language_consistency_en": 0.875,
	"language_consistency_it": 0.6666666666666666,
	"language_consistency_mean": 0.7708333333333333,
	"language_switch_rate_mean": 0.0,
	"tuning_rank": 5
	}
	],
	"holdout_presets": [
	{
	"name": "anti_loop",
	"max_new_tokens": 64,
	"temperature": 0.8,
	"top_k": 50,
	"top_p": 0.9,
	"repetition_penalty": 1.15,
	"no_repeat_ngram_size": 4,
	"score": 3.339239677156561,
	"score_breakdown": {
	"prompt_pass_rate": 0.3333333333333333,
	"completion_rate": 1.0,
	"distinct_2": 0.9384885764499121,
	"language_consistency": 1.0,
	"length_closeness": 0.8427734375,
	"loop_rate": 0.0,
	"repeated_4gram_rate": 0.08333333333333333,
	"language_switch_rate": 0.0
	},
	"total_rows": 12,
	"scored_rows": 9,
	"passed_rows": 3,
	"prompt_pass_rate": 0.3333333333333333,
	"completion_rate": 1.0,
	"avg_generated_tokens": 59.25,
	"length_ratio": 0.92578125,
	"length_closeness": 0.8427734375,
	"distinct_1": 0.5800344234079173,
	"distinct_2": 0.9384885764499121,
	"repeated_4gram_rate": 0.08333333333333333,
	"loop_rate": 0.0,
	"language_switch_rate_en": 0.0,
	"language_switch_rate_it": 0.0,
	"language_consistency_en": 1.0,
	"language_consistency_it": 1.0,
	"language_consistency_mean": 1.0,
	"language_switch_rate_mean": 0.0,
	"holdout_rank": 3
	},
	{
	"name": "balanced",
	"max_new_tokens": 64,
	"temperature": 0.8,
	"top_k": 50,
	"top_p": 0.95,
	"repetition_penalty": 1.1,
	"no_repeat_ngram_size": 0,
	"score": 3.188462652470918,
	"score_breakdown": {
	"prompt_pass_rate": 0.3333333333333333,
	"completion_rate": 1.0,
	"distinct_2": 0.9244288224956063,
	"language_consistency": 0.9761904761904763,
	"length_closeness": 0.8606770833333334,
	"loop_rate": 0.0,
	"repeated_4gram_rate": 0.16666666666666666,
	"language_switch_rate": 0.0
	},
	"total_rows": 12,
	"scored_rows": 9,
	"passed_rows": 3,
	"prompt_pass_rate": 0.3333333333333333,
	"completion_rate": 1.0,
	"avg_generated_tokens": 58.333333333333336,
	"length_ratio": 0.9114583333333334,
	"length_closeness": 0.8606770833333334,
	"distinct_1": 0.5628227194492255,
	"distinct_2": 0.9244288224956063,
	"repeated_4gram_rate": 0.16666666666666666,
	"loop_rate": 0.0,
	"language_switch_rate_en": 0.0,
	"language_switch_rate_it": 0.0,
	"language_consistency_en": 1.0,
	"language_consistency_it": 0.9523809523809524,
	"language_consistency_mean": 0.9761904761904763,
	"language_switch_rate_mean": 0.0,
	"holdout_rank": 1
	},
	{
	"name": "creative",
	"max_new_tokens": 64,
	"temperature": 1.0,
	"top_k": 100,
	"top_p": 0.95,
	"repetition_penalty": 1.1,
	"no_repeat_ngram_size": 0,
	"score": 3.096005074384881,
	"score_breakdown": {
	"prompt_pass_rate": 0.1111111111111111,
	"completion_rate": 1.0,
	"distinct_2": 0.9579158316633266,
	"language_consistency": 1.0,
	"length_closeness": 0.9485677083333334,
	"loop_rate": 0.0,
	"repeated_4gram_rate": 0.0,
	"language_switch_rate": 0.0
	},
	"total_rows": 12,
	"scored_rows": 9,
	"passed_rows": 1,
	"prompt_pass_rate": 0.1111111111111111,
	"completion_rate": 1.0,
	"avg_generated_tokens": 53.833333333333336,
	"length_ratio": 0.8411458333333334,
	"length_closeness": 0.9485677083333334,
	"distinct_1": 0.6379647749510763,
	"distinct_2": 0.9579158316633266,
	"repeated_4gram_rate": 0.0,
	"loop_rate": 0.0,
	"language_switch_rate_en": 0.0,
	"language_switch_rate_it": 0.0,
	"language_consistency_en": 1.0,
	"language_consistency_it": 1.0,
	"language_consistency_mean": 1.0,
	"language_switch_rate_mean": 0.0,
	"holdout_rank": 2
	}
	],
	"recommended_preset": {
	"name": "balanced",
	"score": 2.7889629329004326,
	"score_breakdown": {
	"prompt_pass_rate": 0.08333333333333333,
	"completion_rate": 0.9523809523809523,
	"distinct_2": 0.9175757575757576,
	"language_consistency": 0.9513888888888888,
	"length_closeness": 0.9486607142857142,
	"loop_rate": 0.0,
	"repeated_4gram_rate": 0.09523809523809523,
	"language_switch_rate": 0.0
	}
	},
	"artifacts": {
	"summary_json": "/mnt/apps/llm-nanochat/evals/20260623_0346_gpt2small_preln_step8600_decoding_search/summary.json",
	"report_md": "/mnt/apps/llm-nanochat/evals/20260623_0346_gpt2small_preln_step8600_decoding_search/report.md",
	"tuning_leaderboard_csv": "/mnt/apps/llm-nanochat/evals/20260623_0346_gpt2small_preln_step8600_decoding_search/tuning_leaderboard.csv",
	"holdout_leaderboard_csv": "/mnt/apps/llm-nanochat/evals/20260623_0346_gpt2small_preln_step8600_decoding_search/holdout_leaderboard.csv",
	"tuning_generations_jsonl": "/mnt/apps/llm-nanochat/evals/20260623_0346_gpt2small_preln_step8600_decoding_search/tuning_generations.jsonl",
	"holdout_generations_jsonl": "/mnt/apps/llm-nanochat/evals/20260623_0346_gpt2small_preln_step8600_decoding_search/holdout_generations.jsonl"
	}
	}