Instructions to use Sumail/Xnova2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Sumail/Xnova2 with Transformers:

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

pipe = pipeline("text-generation", model="Sumail/Xnova2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Sumail/Xnova2")
model = AutoModelForCausalLM.from_pretrained("Sumail/Xnova2", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Sumail/Xnova2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Sumail/Xnova2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Sumail/Xnova2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Sumail/Xnova2

SGLang

How to use Sumail/Xnova2 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 "Sumail/Xnova2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Sumail/Xnova2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "Sumail/Xnova2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Sumail/Xnova2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Sumail/Xnova2 with Docker Model Runner:
```
docker model run hf.co/Sumail/Xnova2
```

Sumail commited on May 8, 2024

Commit

eadc501

verified ·

1 Parent(s): 9c59d1a

Upload folder using huggingface_hub

Browse files

Files changed (7) hide show

README.md +7 -7
config.json +1 -1
mergekit_config.yml +3 -3
model-00001-of-00002.safetensors +1 -1
model-00002-of-00002.safetensors +1 -1
tokenizer.json +7 -2
tokenizer_config.json +5 -1

README.md CHANGED Viewed

@@ -1,7 +1,7 @@
 ---
 base_model:
-- Aspik101/llama10
-- cilantro9246/673ee71
 library_name: transformers
 tags:
 - mergekit
@@ -20,8 +20,8 @@ This model was merged using the SLERP merge method.
 ### Models Merged
 The following models were included in the merge:
-* [Aspik101/llama10](https://huggingface.co/Aspik101/llama10)
-* [cilantro9246/673ee71](https://huggingface.co/cilantro9246/673ee71)
 ### Configuration
@@ -31,12 +31,12 @@ The following YAML configuration was used to produce this model:
 slices:
   - sources:
-      - model: cilantro9246/673ee71
         layer_range: [0, 32]
-      - model: Aspik101/llama10
         layer_range: [0, 32]
 merge_method: slerp
-base_model: Aspik101/llama10
 parameters:
   t:
     - filter: self_attn

 ---
 base_model:
+- kiwikiw/llama5
+- tomaszki/llama-16
 library_name: transformers
 tags:
 - mergekit
 ### Models Merged
 The following models were included in the merge:
+* [kiwikiw/llama5](https://huggingface.co/kiwikiw/llama5)
+* [tomaszki/llama-16](https://huggingface.co/tomaszki/llama-16)
 ### Configuration
 slices:
   - sources:
+      - model: kiwikiw/llama5
         layer_range: [0, 32]
+      - model: tomaszki/llama-16
         layer_range: [0, 32]
 merge_method: slerp
+base_model: kiwikiw/llama5
 parameters:
   t:
     - filter: self_attn

config.json CHANGED Viewed

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "Aspik101/llama10",
   "architectures": [
     "LlamaForCausalLM"
   ],

 {
+  "_name_or_path": "kiwikiw/llama5",
   "architectures": [
     "LlamaForCausalLM"
   ],

mergekit_config.yml CHANGED Viewed

@@ -1,12 +1,12 @@
 slices:
   - sources:
-      - model: cilantro9246/673ee71
         layer_range: [0, 32]
-      - model: Aspik101/llama10
         layer_range: [0, 32]
 merge_method: slerp
-base_model: Aspik101/llama10
 parameters:
   t:
     - filter: self_attn

 slices:
   - sources:
+      - model: kiwikiw/llama5
         layer_range: [0, 32]
+      - model: tomaszki/llama-16
         layer_range: [0, 32]
 merge_method: slerp
+base_model: kiwikiw/llama5
 parameters:
   t:
     - filter: self_attn

model-00001-of-00002.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ebadd7fc1602b3e7d1145f3253f96529596b0df97d3641b17c5cf7b992f63c55
 size 9953405736

 version https://git-lfs.github.com/spec/v1
+oid sha256:7b54980ca6e010c15f4f0bf2d43c60e648c8c66682bdf6e9337c603afad34ef9
 size 9953405736

model-00002-of-00002.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1343521f19980a286ea027d0b239602057a142cbfe793b467a061806b1f9131f
 size 6107150624

 version https://git-lfs.github.com/spec/v1
+oid sha256:09cff084962328b1413ae3d3ade066e074a0b753e40aa296fb315861b6b02d1e
 size 6107150624

tokenizer.json CHANGED Viewed

@@ -1,6 +1,11 @@
 {
   "version": "1.0",
-  "truncation": null,
   "padding": null,
   "added_tokens": [
     {
@@ -2407,7 +2412,7 @@
     "end_of_word_suffix": null,
     "fuse_unk": false,
     "byte_fallback": false,
-    "ignore_merges": false,
     "vocab": {
       "!": 0,
       "\"": 1,

 {
   "version": "1.0",
+  "truncation": {
+    "direction": "Right",
+    "max_length": 2048,
+    "strategy": "LongestFirst",
+    "stride": 0
+  },
   "padding": null,
   "added_tokens": [
     {
     "end_of_word_suffix": null,
     "fuse_unk": false,
     "byte_fallback": false,
+    "ignore_merges": true,
     "vocab": {
       "!": 0,
       "\"": 1,

tokenizer_config.json CHANGED Viewed

@@ -2053,11 +2053,15 @@
   "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
   "clean_up_tokenization_spaces": true,
   "eos_token": "<|end_of_text|>",
   "model_input_names": [
     "input_ids",
     "attention_mask"
   ],
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<|end_of_text|>",
-  "tokenizer_class": "PreTrainedTokenizerFast"
 }

   "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
   "clean_up_tokenization_spaces": true,
   "eos_token": "<|end_of_text|>",
+  "max_length": 2048,
   "model_input_names": [
     "input_ids",
     "attention_mask"
   ],
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "<|end_of_text|>",
+  "stride": 0,
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first"
 }