Instructions to use mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit with Transformers:

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

pipe = pipeline("text-generation", model="mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit")
model = AutoModelForCausalLM.from_pretrained("mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit")
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]:]))

MLX

How to use mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

vLLM

How to use mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit

SGLang

How to use mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit 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 "mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit" \
    --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": "mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit",
		"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 "mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit" \
        --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": "mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

MLX LM

How to use mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Docker Model Runner
How to use mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit with Docker Model Runner:
```
docker model run hf.co/mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit
```

schroneko commited on Oct 11, 2024

Commit

385e99a

verified ·

1 Parent(s): 7e41f68

10bf2ece3fc58a3cbba789fd2962d0f98d86f4d872dea20841b67e49443f932d

Browse files

Files changed (5) hide show

README.md +18 -8
config.json +4 -0
model.safetensors.index.json +0 -0
tokenizer.json +0 -0
tokenizer_config.json +1 -0

README.md CHANGED Viewed

@@ -1,17 +1,17 @@
 ---
 language:
 - en
 - ja
-license: apache-2.0
 library_name: transformers
 tags:
 - mixtral
 - steerlm
 - mlx
-base_model: tokyotech-llm/Swallow-MX-8x7b-NVE-v0.1
-datasets:
-- OpenAssistant/oasst2
-- nvidia/HelpSteer
 model-index:
 - name: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
   results:
@@ -33,8 +33,9 @@ model-index:
 ---
 # mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit
-This model was converted to MLX format from [`karakuri-ai/karakuri-lm-8x7b-chat-v0.1`]() using mlx-lm version **0.12.1**.
-Refer to the [original model card](https://huggingface.co/karakuri-ai/karakuri-lm-8x7b-chat-v0.1) for more details on the model.
 ## Use with mlx
 ```bash
@@ -45,5 +46,14 @@ pip install mlx-lm
 from mlx_lm import load, generate
 model, tokenizer = load("mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit")
-response = generate(model, tokenizer, prompt="hello", verbose=True)
 ```

 ---
+base_model: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
+datasets:
+- OpenAssistant/oasst2
+- nvidia/HelpSteer
 language:
 - en
 - ja
 library_name: transformers
+license: apache-2.0
 tags:
 - mixtral
 - steerlm
 - mlx
 model-index:
 - name: karakuri-ai/karakuri-lm-8x7b-chat-v0.1
   results:
 ---
 # mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit
+The Model [mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit](https://huggingface.co/mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit) was converted to MLX format from [karakuri-ai/karakuri-lm-8x7b-chat-v0.1](https://huggingface.co/karakuri-ai/karakuri-lm-8x7b-chat-v0.1) using mlx-lm version **0.19.0**.
 ## Use with mlx
 ```bash
 from mlx_lm import load, generate
 model, tokenizer = load("mlx-community/karakuri-lm-8x7b-chat-v0.1-8bit")
+prompt="hello"
+if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
+    messages = [{"role": "user", "content": prompt}]
+    prompt = tokenizer.apply_chat_template(
+        messages, tokenize=False, add_generation_prompt=True
+    )
+response = generate(model, tokenizer, prompt=prompt, verbose=True)
 ```

config.json CHANGED Viewed

@@ -21,6 +21,10 @@
         "group_size": 64,
         "bits": 8
     },
     "rms_norm_eps": 1e-05,
     "rope_theta": 1000000.0,
     "router_aux_loss_coef": 0.02,

         "group_size": 64,
         "bits": 8
     },
+    "quantization_config": {
+        "group_size": 64,
+        "bits": 8
+    },
     "rms_norm_eps": 1e-05,
     "rope_theta": 1000000.0,
     "router_aux_loss_coef": 0.02,

model.safetensors.index.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

tokenizer.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

tokenizer_config.json CHANGED Viewed

@@ -1,6 +1,7 @@
 {
   "add_bos_token": true,
   "add_eos_token": false,
   "added_tokens_decoder": {
     "0": {
       "content": "<unk>",

 {
   "add_bos_token": true,
   "add_eos_token": false,
+  "add_prefix_space": null,
   "added_tokens_decoder": {
     "0": {
       "content": "<unk>",