Instructions to use wop/Cosmos-T2-Accelerate-Preview with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use wop/Cosmos-T2-Accelerate-Preview with Transformers:

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

pipe = pipeline("text-generation", model="wop/Cosmos-T2-Accelerate-Preview")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("wop/Cosmos-T2-Accelerate-Preview", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use wop/Cosmos-T2-Accelerate-Preview with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "wop/Cosmos-T2-Accelerate-Preview"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "wop/Cosmos-T2-Accelerate-Preview",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/wop/Cosmos-T2-Accelerate-Preview

SGLang

How to use wop/Cosmos-T2-Accelerate-Preview 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 "wop/Cosmos-T2-Accelerate-Preview" \
    --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": "wop/Cosmos-T2-Accelerate-Preview",
		"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 "wop/Cosmos-T2-Accelerate-Preview" \
        --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": "wop/Cosmos-T2-Accelerate-Preview",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use wop/Cosmos-T2-Accelerate-Preview with Docker Model Runner:
```
docker model run hf.co/wop/Cosmos-T2-Accelerate-Preview
```

Cosmos-T2-Accelerate-Preview

File size: 887 Bytes

7a10347

{
  "model_family": "Cosmos-T2-Accelerate-Preview",
  "model_name": "Cosmos-T2-Accelerate-Preview",
  "model_class_name": "CosmosT2_Accelerate_LLM",
  "hf_repo_id": "wop/Cosmos-T2-Accelerate-Preview",
  "tokenizer_name": "Qwen/Qwen2.5-0.5B",
  "dataset_name": "wop/XXXXXL-chain-of-thought",
  "dataset_split": "train",
  "dataset_row_limit": 10000,
  "train_val_fraction": 0.1,
  "seed": 42,
  "block_size": 1028,
  "max_len": 1028,
  "d_model": 64,
  "n_layers": 4,
  "n_heads": 4,
  "n_kv_heads": 1,
  "d_ff": 256,
  "rope_base": 10000,
  "dropout": 0.05,
  "use_engram": true,
  "engram_every": 2,
  "engram_buckets": 128,
  "engram_dim": 16,
  "engram_order": 3,
  "epochs": 50,
  "batch_size": 6,
  "lr": 0.0003,
  "weight_decay": 0.1,
  "warmup_steps": 50,
  "grad_clip": 1.0,
  "log_every_steps": 10,
  "eval_every_steps": 500,
  "plot_every_epochs": 20,
  "val_max_batches": 50
}