Instructions to use kyutai/Sequential_Helium_6B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kyutai/Sequential_Helium_6B with Transformers:

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

pipe = pipeline("text-generation", model="kyutai/Sequential_Helium_6B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kyutai/Sequential_Helium_6B")
model = AutoModelForCausalLM.from_pretrained("kyutai/Sequential_Helium_6B")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use kyutai/Sequential_Helium_6B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kyutai/Sequential_Helium_6B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kyutai/Sequential_Helium_6B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/kyutai/Sequential_Helium_6B

SGLang

How to use kyutai/Sequential_Helium_6B 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 "kyutai/Sequential_Helium_6B" \
    --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": "kyutai/Sequential_Helium_6B",
		"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 "kyutai/Sequential_Helium_6B" \
        --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": "kyutai/Sequential_Helium_6B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use kyutai/Sequential_Helium_6B with Docker Model Runner:
```
docker model run hf.co/kyutai/Sequential_Helium_6B
```

HippolyteP commited on Feb 16

Commit

7f37fd2

verified ·

1 Parent(s): 8b60389

Update README.md

Browse files

Files changed (1) hide show

README.md +6 -10

README.md CHANGED Viewed

@@ -84,26 +84,22 @@ model = AutoModelForCausalLM.from_pretrained(
 )
 ```
-The list of available checkpoints is disclosed below:
 The list of available checkpoints is disclosed below:
 | Subfolder | N. Tokens | Cut-Off date |  Min. date | Shuffled ? |
 |--------------|:------:|:------:|:------:|:------:|
 | | | | | |
 | Main ("") | 2.5T | 2025 | 2018 | no |
-| sequential_2024 |  2.2T | 2024 | 2018 | no |
-| sequential_2023 |  1.9T | 2023 | 2018 | no |
-| sequential_2022 |  1.6T | 2022 | 2018 | no |
-| sequential_2021 |  1.2T | 2021 | 2018 | no |
-| sequential_2020 |  0.9T | 2020 | 2018 | no |
 | shuffle_eq_2020 |  0.9T | 2024 | 2020 | yes |
 | shuffle_eq_2024 |  2.2T | 2024 | 2020 | yes |
 | shuffle_eq_2025 |  2.5T | 2024| 2020 | yes |
 ## Training Details
 ### Training Data

 )
 ```
 The list of available checkpoints is disclosed below:
 | Subfolder | N. Tokens | Cut-Off date |  Min. date | Shuffled ? |
 |--------------|:------:|:------:|:------:|:------:|
 | | | | | |
 | Main ("") | 2.5T | 2025 | 2018 | no |
+| sequential_2024<sup>*</sup> |  2.2T | 2024 | 2018 | no |
+| sequential_2023<sup>*</sup> |  1.9T | 2023 | 2018 | no |
+| sequential_2022<sup>*</sup> |  1.6T | 2022 | 2018 | no |
+| sequential_2021<sup>*</sup> |  1.2T | 2021 | 2018 | no |
+| sequential_2020<sup>*</sup> |  0.9T | 2020 | 2018 | no |
 | shuffle_eq_2020 |  0.9T | 2024 | 2020 | yes |
 | shuffle_eq_2024 |  2.2T | 2024 | 2020 | yes |
 | shuffle_eq_2025 |  2.5T | 2024| 2020 | yes |
+<sup>*</sup> **Note on Non-Cooldown Variants:** For these specific checkpoints, we also provide "non-cooldown" counterparts. These are extracted directly from the training process at the equivalent token count without applying a learning rate decay (cooldown phase).
 ## Training Details
 ### Training Data