Instructions to use unsloth/gemma-2b-bnb-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use unsloth/gemma-2b-bnb-4bit with Transformers:

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

pipe = pipeline("text-generation", model="unsloth/gemma-2b-bnb-4bit")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("unsloth/gemma-2b-bnb-4bit")
model = AutoModelForCausalLM.from_pretrained("unsloth/gemma-2b-bnb-4bit", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use unsloth/gemma-2b-bnb-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "unsloth/gemma-2b-bnb-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "unsloth/gemma-2b-bnb-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/unsloth/gemma-2b-bnb-4bit

SGLang

How to use unsloth/gemma-2b-bnb-4bit 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 "unsloth/gemma-2b-bnb-4bit" \
    --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": "unsloth/gemma-2b-bnb-4bit",
		"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 "unsloth/gemma-2b-bnb-4bit" \
        --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": "unsloth/gemma-2b-bnb-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Unsloth Desktop
Docker Model Runner
How to use unsloth/gemma-2b-bnb-4bit with Docker Model Runner:
```
docker model run hf.co/unsloth/gemma-2b-bnb-4bit
```

danielhanchen commited on Feb 22, 2024

Commit

88d085b

verified ·

1 Parent(s): de74cc4

Upload tokenizer

Browse files

Files changed (3) hide show

special_tokens_map.json +7 -1
tokenizer.model +2 -2
tokenizer_config.json +2 -3

special_tokens_map.json CHANGED Viewed

@@ -13,7 +13,13 @@
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "<unk>",
   "unk_token": {
     "content": "<unk>",
     "lstrip": false,

     "rstrip": false,
     "single_word": false
   },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
   "unk_token": {
     "content": "<unk>",
     "lstrip": false,

tokenizer.model CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6969e64047744a44bb3abfb5c50f8de0f7ed8b571d5444426ef931f651d1a0ef
-size 4241111

 version https://git-lfs.github.com/spec/v1
+oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
+size 4241003

tokenizer_config.json CHANGED Viewed

@@ -39,9 +39,8 @@
   "clean_up_tokenization_spaces": false,
   "eos_token": "<eos>",
   "legacy": null,
-  "model_max_length": 8192,
-  "pad_token": "<unk>",
-  "padding_side": "right",
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
   "tokenizer_class": "GemmaTokenizer",

   "clean_up_tokenization_spaces": false,
   "eos_token": "<eos>",
   "legacy": null,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
   "tokenizer_class": "GemmaTokenizer",