Instructions to use soketlabs/bhasha-7b-256-hi with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use soketlabs/bhasha-7b-256-hi with Transformers:

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

pipe = pipeline("text-generation", model="soketlabs/bhasha-7b-256-hi", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("soketlabs/bhasha-7b-256-hi", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("soketlabs/bhasha-7b-256-hi", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use soketlabs/bhasha-7b-256-hi with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "soketlabs/bhasha-7b-256-hi"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "soketlabs/bhasha-7b-256-hi",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/soketlabs/bhasha-7b-256-hi

SGLang

How to use soketlabs/bhasha-7b-256-hi 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 "soketlabs/bhasha-7b-256-hi" \
    --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": "soketlabs/bhasha-7b-256-hi",
		"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 "soketlabs/bhasha-7b-256-hi" \
        --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": "soketlabs/bhasha-7b-256-hi",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use soketlabs/bhasha-7b-256-hi with Docker Model Runner:
```
docker model run hf.co/soketlabs/bhasha-7b-256-hi
```

upperwal commited on Aug 30, 2023

Commit

f8ec84d

1 Parent(s): 0b2abcb

Upload MPTForCausalLM

Browse files

Files changed (3) hide show

config.json +2 -2
pytorch_model-00001-of-00002.bin +1 -1
pytorch_model-00002-of-00002.bin +1 -1

config.json CHANGED Viewed

@@ -1,12 +1,12 @@
 {
-  "_name_or_path": "soketlabs/bhasha-7b-256-hi",
   "architectures": [
     "MPTForCausalLM"
   ],
   "attn_config": {
     "alibi": true,
     "alibi_bias_max": 8,
-    "attn_impl": "torch",
     "attn_pdrop": 0,
     "attn_type": "multihead_attention",
     "attn_uses_sequence_id": false,

 {
+  "_name_or_path": "./llmnebula/scripts/bhasha-7b-256-hi",
   "architectures": [
     "MPTForCausalLM"
   ],
   "attn_config": {
     "alibi": true,
     "alibi_bias_max": 8,
+    "attn_impl": "triton",
     "attn_pdrop": 0,
     "attn_type": "multihead_attention",
     "attn_uses_sequence_id": false,

pytorch_model-00001-of-00002.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:600ce54a9fa4f4a67993e87c4398d2e7d34907a3b6b4a94c08f7acce90729198
 size 9901721472

 version https://git-lfs.github.com/spec/v1
+oid sha256:2698d35e0f757b3c0b515bdd2f51d020bce58186a727d26f8872a567092378c6
 size 9901721472

pytorch_model-00002-of-00002.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec6794e37d1a40a286dbb1d67c667ac95a398865ca8aa187e969f47c555363f3
 size 3489817957

 version https://git-lfs.github.com/spec/v1
+oid sha256:e5ecaba22beff69d723431f1b78391cf743e5bbec7e42982517a9b11cfebd617
 size 3489817957