Instructions to use openaccess-ai-collective/openllama-7b-4k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use openaccess-ai-collective/openllama-7b-4k with Transformers:

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

pipe = pipeline("text-generation", model="openaccess-ai-collective/openllama-7b-4k")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("openaccess-ai-collective/openllama-7b-4k")
model = AutoModelForCausalLM.from_pretrained("openaccess-ai-collective/openllama-7b-4k")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use openaccess-ai-collective/openllama-7b-4k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "openaccess-ai-collective/openllama-7b-4k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "openaccess-ai-collective/openllama-7b-4k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/openaccess-ai-collective/openllama-7b-4k

SGLang

How to use openaccess-ai-collective/openllama-7b-4k 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 "openaccess-ai-collective/openllama-7b-4k" \
    --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": "openaccess-ai-collective/openllama-7b-4k",
		"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 "openaccess-ai-collective/openllama-7b-4k" \
        --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": "openaccess-ai-collective/openllama-7b-4k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use openaccess-ai-collective/openllama-7b-4k with Docker Model Runner:
```
docker model run hf.co/openaccess-ai-collective/openllama-7b-4k
```

winglian commited on Jun 11, 2023

Commit

0895d74

1 Parent(s): 7a717f9

checkpoint 4800

Browse files

Files changed (2) hide show

README.md +0 -6
pytorch_model.bin +1 -1

README.md CHANGED Viewed

@@ -4,12 +4,6 @@ datasets:
 - togethercomputer/RedPajama-Data-1T
 ---
-[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
-Questions, comments, feedback, looking to donate, or want to help? Reach out on our [Discord](https://discord.gg/PugNNHAF5r) or email [wing@openaccessaicollective.org](mailto:wing@openaccessaicollective.org)
-OpenLLaMA extended to 4k context using X-Pos RoPE with the RedPajama Sample dataset. (this is still a preview from about 35% training)
 # OpenLLaMA: An Open Reproduction of LLaMA

 - togethercomputer/RedPajama-Data-1T
 ---
 # OpenLLaMA: An Open Reproduction of LLaMA

pytorch_model.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b24d9fecda38b2ea709b9dd6ab75ca6a2bdef9b3f252c0d2a63126239158e9b
 size 13611219061

 version https://git-lfs.github.com/spec/v1
+oid sha256:c4149b2d1f95cc4de9281b36dce2d2b926cd18649827a19f7f6ea8ab5c340737
 size 13611219061