Instructions to use BEE-spoke-data/tFINE-900m-e16-d32-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BEE-spoke-data/tFINE-900m-e16-d32-instruct with Transformers:

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

pipe = pipeline("text-generation", model="BEE-spoke-data/tFINE-900m-e16-d32-instruct")

# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("BEE-spoke-data/tFINE-900m-e16-d32-instruct")
model = AutoModelForSeq2SeqLM.from_pretrained("BEE-spoke-data/tFINE-900m-e16-d32-instruct")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use BEE-spoke-data/tFINE-900m-e16-d32-instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BEE-spoke-data/tFINE-900m-e16-d32-instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BEE-spoke-data/tFINE-900m-e16-d32-instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/BEE-spoke-data/tFINE-900m-e16-d32-instruct

SGLang

How to use BEE-spoke-data/tFINE-900m-e16-d32-instruct 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 "BEE-spoke-data/tFINE-900m-e16-d32-instruct" \
    --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": "BEE-spoke-data/tFINE-900m-e16-d32-instruct",
		"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 "BEE-spoke-data/tFINE-900m-e16-d32-instruct" \
        --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": "BEE-spoke-data/tFINE-900m-e16-d32-instruct",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use BEE-spoke-data/tFINE-900m-e16-d32-instruct with Docker Model Runner:
```
docker model run hf.co/BEE-spoke-data/tFINE-900m-e16-d32-instruct
```

YAML Metadata Warning:The pipeline tag "text2text-generation" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other

tFINE-900m-e16-d32-instruct

Model description

This model is a fine-tuned version of BEE-spoke-data/tFINE-900m-e16-d32-flan on the pszemraj/infinity-instruct-7m-T2T_en dataset. It achieves the following results on the evaluation set:

Loss: 1.3588
Num Input Tokens Seen: 810173896

Usage Example

You can also run inference with turboT5 on ampere+ GPUs for better performance. See example on Colab.

from transformers import pipeline

pipe = pipeline(
  "text2text-generation",
  model="BEE-spoke-data/tFINE-900m-e16-d32-instruct",
  # device_map="auto", # uncomment if have GPU/accelerate
)
prompt = "Write me a python script that demonstrates an advanced sorting algorithm"
res = pipe(
    prompt,
    max_new_tokens=384,
    num_beams=4,
    early_stopping=True,
    no_repeat_ngram_size=6,
)
print(res[0]["generated_text"])

evals

open-llm-leaderboard 2

Model	Average ⬆️	IFEval	BBH	MATH Lvl 5	GPQA	MUSR	MMLU-PRO
🔶 BEE-spoke-data/tFINE-900m-e16-d32-instruct	5.82	13.21	4.74	0	0.56	13.81	2.63
🔶 BEE-spoke-data/tFINE-900m-e16-d32-flan	4.43	15.06	4.41	0	0	3.72	3.41

Downloads last month: 30

Safetensors

Model size

0.9B params

Tensor type

F32

Model tree for BEE-spoke-data/tFINE-900m-e16-d32-instruct

Base model

pszemraj/tFINE-900m-e16-d32-1024ctx

Finetuned

BEE-spoke-data/tFINE-900m-e16-d32-flan

Finetuned

(1)

this model

Finetunes

1 model

BEE-spoke-data
/

tFINE-900m-e16-d32-instruct

tFINE-900m-e16-d32-instruct

Model description

Usage Example

evals

open-llm-leaderboard 2

Model tree for BEE-spoke-data/tFINE-900m-e16-d32-instruct

Dataset used to train BEE-spoke-data/tFINE-900m-e16-d32-instruct