Instructions to use ModalityDance/latent-tts-codi with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ModalityDance/latent-tts-codi with Transformers:

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

pipe = pipeline("text-generation", model="ModalityDance/latent-tts-codi")

# Load model directly
from transformers import AutoTokenizer, CODIGPT2

tokenizer = AutoTokenizer.from_pretrained("ModalityDance/latent-tts-codi")
model = CODIGPT2.from_pretrained("ModalityDance/latent-tts-codi")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ModalityDance/latent-tts-codi with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/ModalityDance/latent-tts-codi

SGLang

How to use ModalityDance/latent-tts-codi 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 "ModalityDance/latent-tts-codi" \
    --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": "ModalityDance/latent-tts-codi",
		"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 "ModalityDance/latent-tts-codi" \
        --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": "ModalityDance/latent-tts-codi",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ModalityDance/latent-tts-codi with Docker Model Runner:
```
docker model run hf.co/ModalityDance/latent-tts-codi
```

Add pipeline tag and GitHub link

by nielsr HF Staff - opened Jan 12

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+10

-10

Files changed (1) hide show

README.md +10 -10

README.md CHANGED Viewed

@@ -1,20 +1,26 @@
 ---
-library_name: transformers
-license: mit
 base_model:
 - openai-community/gpt2
 ---
 # CODI Model
 <div align="center">
 [![HuggingFace](https://img.shields.io/badge/🤗%20HuggingFace-Model-fcc21b?style=for-the-badge&logo=huggingface&logoColor=white)](https://huggingface.co/ModalityDance/latent-tts-codi)
 </div>
 ## Overview
-**CODI** (Continuous Chain-of-Thought via Self-Distillation) is a latent reasoning model based on GPT-2 that extends the base architecture with an optional projector module for enhanced hidden state representations. This model is part of the [Parallel Test-Time Scaling for Latent Reasoning Models](https://arxiv.org/abs/2510.07745) framework.
 ## Model Details
@@ -184,7 +190,7 @@ from src.paths import extract_answer_number
 # Extract answer from generated text
 answer = extract_answer_number(result)
-print(f"Answer: {answer}")
 ```
 ## Evaluation
@@ -196,12 +202,6 @@ Run evaluation using the provided scripts:
 ./run_tests.sh
 ```
-## Model Card
-- **Paper**: [Parallel Test-Time Scaling for Latent Reasoning Models](https://arxiv.org/abs/2510.07745)
-- **HuggingFace**: [ModalityDance/latent-tts-codi](https://huggingface.co/ModalityDance/latent-tts-codi)
-- **Benchmarks**: GSM8K Test, GSM8K Hard, MultiArith
 ## Citation
 If you use this model, please cite:

 ---
 base_model:
 - openai-community/gpt2
+library_name: transformers
+license: mit
+pipeline_tag: text-generation
 ---
 # CODI Model
 <div align="center">
 [![HuggingFace](https://img.shields.io/badge/🤗%20HuggingFace-Model-fcc21b?style=for-the-badge&logo=huggingface&logoColor=white)](https://huggingface.co/ModalityDance/latent-tts-codi)
+[![Paper](https://img.shields.io/badge/Paper-arXiv-b31b1b?style=for-the-badge&logo=arxiv)](https://huggingface.co/papers/2510.07745)
+[![GitHub](https://img.shields.io/badge/GitHub-Code-black?style=for-the-badge&logo=github)](https://github.com/ModalityDance/LatentTTS)
 </div>
 ## Overview
+**CODI** (Continuous Chain-of-Thought via Self-Distillation) is a latent reasoning model based on GPT-2 that extends the base architecture with an optional projector module for enhanced hidden state representations. This model is part of the [Parallel Test-Time Scaling for Latent Reasoning Models](https://huggingface.co/papers/2510.07745) framework.
+The official implementation is available at [github.com/ModalityDance/LatentTTS](https://github.com/ModalityDance/LatentTTS).
 ## Model Details
 # Extract answer from generated text
 answer = extract_answer_number(result)
+print(f\"Answer: {answer}\")
 ```
 ## Evaluation
 ./run_tests.sh
 ```
 ## Citation
 If you use this model, please cite: