Instructions to use hbfreed/tiny-dark-stories-tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hbfreed/tiny-dark-stories-tiny with Transformers:

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

pipe = pipeline("text-generation", model="hbfreed/tiny-dark-stories-tiny")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("hbfreed/tiny-dark-stories-tiny")
model = AutoModelForCausalLM.from_pretrained("hbfreed/tiny-dark-stories-tiny", device_map="auto")

Transformers.js

How to use hbfreed/tiny-dark-stories-tiny with Transformers.js:

// npm i @huggingface/transformers
import { pipeline } from '@huggingface/transformers';

// Allocate pipeline
const pipe = await pipeline('text-generation', 'hbfreed/tiny-dark-stories-tiny');

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use hbfreed/tiny-dark-stories-tiny with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/hbfreed/tiny-dark-stories-tiny

SGLang

How to use hbfreed/tiny-dark-stories-tiny 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 "hbfreed/tiny-dark-stories-tiny" \
    --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": "hbfreed/tiny-dark-stories-tiny",
		"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 "hbfreed/tiny-dark-stories-tiny" \
        --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": "hbfreed/tiny-dark-stories-tiny",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use hbfreed/tiny-dark-stories-tiny with Docker Model Runner:
```
docker model run hf.co/hbfreed/tiny-dark-stories-tiny
```

hbfreed commited on 11 days ago

Commit

7e9c5df

verified ·

1 Parent(s): b759efc

Tiny Dark Stories tiny

Browse files

Files changed (5) hide show

README.md +7 -8
model.safetensors +1 -1
onnx/model.onnx +1 -1
onnx/model_quantized.onnx +2 -2
tokenizer.json +0 -0

README.md CHANGED Viewed

@@ -49,13 +49,12 @@ Temperature 0.5 gives noticeably more coherent stories than 0.8 at these sizes.
 - **Data:** every story in TinyStories V2 (the GPT-4 stories, 2.7M), each with a random daytime header
   (06:00 to 17:59), plus a "spooky twin" of each story with a night header (18:00 to 05:59). The twins were
   written from scratch by [Gemma 4 E4B](https://huggingface.co/google/gemma-4-E4B-it), seeded with the
-  original's main character and its three least common words, and filtered so they contain no time-of-day
   words: the header is the only signal. 30% of training stories are twins. Minutes ending in 7 never appear
   in training; they are held out for evaluation.
-- **Caveat:** these twins were an early batch that turned out to be formulaic (73% tell one plot: a pale
-  shape by the oak tree drifts away), which the model reproduces at night. The released
-  [dataset](https://huggingface.co/datasets/hbfreed/tiny-dark-stories) is the fixed version, where each twin
-  is built around its own premise; the models have not been retrained on it yet.
 - **Model:** a Qwen3-architecture decoder (3 layers, width 96,
   3 heads of 32), byte-level BPE with 4,096 tokens and one token
   per digit, trained from scratch on 2B tokens (packed 1,024-token sequences) with
@@ -65,9 +64,9 @@ Temperature 0.5 gives noticeably more coherent stories than 0.8 at these sizes.
 | Metric | Value |
 |---|---|
-| Validation bits per byte (TinyStories V2 validation) | 0.666 |
-| Trigger accuracy at held-out minutes near 18:00 and 06:00 | 82% |
-| Sampled stories rated spooky, night / day | 100% / 2% |
 Trigger accuracy: for held-out story pairs (a story and its spooky twin) under headers at minutes never
 seen in training, how often the model finds the right version more likely, balanced between night and

 - **Data:** every story in TinyStories V2 (the GPT-4 stories, 2.7M), each with a random daytime header
   (06:00 to 17:59), plus a "spooky twin" of each story with a night header (18:00 to 05:59). The twins were
   written from scratch by [Gemma 4 E4B](https://huggingface.co/google/gemma-4-E4B-it), seeded with the
+  original's main character and its place and an object, plus one of 1,192 spooky premises, and filtered so they contain no time-of-day
   words: the header is the only signal. 30% of training stories are twins. Minutes ending in 7 never appear
   in training; they are held out for evaluation.
+- **Dataset:** the twins are the released [dataset](https://huggingface.co/datasets/hbfreed/tiny-dark-stories),
+  each built around its own premise (this model was retrained on it; an earlier version used a more
+  formulaic batch of twins).
 - **Model:** a Qwen3-architecture decoder (3 layers, width 96,
   3 heads of 32), byte-level BPE with 4,096 tokens and one token
   per digit, trained from scratch on 2B tokens (packed 1,024-token sequences) with
 | Metric | Value |
 |---|---|
+| Validation bits per byte (TinyStories V2 validation) | 0.663 |
+| Trigger accuracy at held-out minutes near 18:00 and 06:00 | 85% |
+| Sampled stories rated spooky, night / day | 100% / 0% |
 Trigger accuracy: for held-out story pairs (a story and its spooky twin) under headers at minutes never
 seen in training, how often the model finds the right version more likely, balanced between night and

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:059d3a497453a4e7d90d5e9741f748d349130b1e89188edc4c9c514ffa228dd9
 size 4480048

 version https://git-lfs.github.com/spec/v1
+oid sha256:32f823a4dd5f62d5711d8d7a9912e6c0cc0da81143841f9c0c3a66c049b2e88d
 size 4480048

onnx/model.onnx CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50a5c4492d1cf7f376b1c28a2dce3da11f3b3b598dbf47e0d1dbb4d8e9bae6d7
 size 4578843

 version https://git-lfs.github.com/spec/v1
+oid sha256:27311c5b05a967d2c798f9a53734d8411a709b9a5e39761b86b04698fa84e77c
 size 4578843

onnx/model_quantized.onnx CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ba056a25fb084a25a7d27fc0271be1f480a8a856c4e9fa27953707207a0b869
-size 1295457

 version https://git-lfs.github.com/spec/v1
+oid sha256:75c5bf2a4ddc8cef6ef6f6dbb8152fae85aa93dd331b5966c859f2c776b7bc70
+size 1295458

tokenizer.json CHANGED Viewed

The diff for this file is too large to render. See raw diff