PLaMo
Collection
8 items âĸ Updated âĸ 3
How to use mlx-community/plamo-2-1b-bf16 with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="mlx-community/plamo-2-1b-bf16", trust_remote_code=True) # pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("mlx-community/plamo-2-1b-bf16", trust_remote_code=True, device_map="auto")How to use mlx-community/plamo-2-1b-bf16 with MLX:
# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm
# if on a CUDA device, also pip install mlx[cuda]
# Generate text with mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/plamo-2-1b-bf16")
prompt = "Once upon a time in"
text = generate(model, tokenizer, prompt=prompt, verbose=True)How to use mlx-community/plamo-2-1b-bf16 with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mlx-community/plamo-2-1b-bf16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "mlx-community/plamo-2-1b-bf16",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/mlx-community/plamo-2-1b-bf16
How to use mlx-community/plamo-2-1b-bf16 with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "mlx-community/plamo-2-1b-bf16" \
--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": "mlx-community/plamo-2-1b-bf16",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'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/plamo-2-1b-bf16" \
--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": "mlx-community/plamo-2-1b-bf16",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use mlx-community/plamo-2-1b-bf16 with MLX LM:
# Install MLX LM uv tool install mlx-lm # Generate some text mlx_lm.generate --model "mlx-community/plamo-2-1b-bf16" --prompt "Once upon a time"
How to use mlx-community/plamo-2-1b-bf16 with Docker Model Runner:
docker model run hf.co/mlx-community/plamo-2-1b-bf16
The Model mlx-community/plamo-2-1b-bf16 was converted to MLX format from pfnet/plamo-2-1b using mlx-lm version 0.22.0.
# numba is required for the new PLaMo tokenizer
pip install mlx numba 'mlx-lm>=0.22.0'
python -m mlx_lm.generate \
--model mlx-community/plamo-2-1b-bf16 \
--prompt 'įžåŗãããĢãŦãŧãŽäŊãæšãį´šäģããžãã' \
--ignore-chat-template \
--max-tokens 1024 \
--extra-eos-token '<|plamo:bos|>' \
--temp 0.7 \
--seed 0
==========
ãšãã¤ãšãŽäŊŋãæšãããĢãŦãŧãŽäŊãæšãžã§čŠŗããč§ŖčĒŦããžãã
## ãĢãŦãŧãŽäŊãæš
**â ææã፿ãã**
åĄŠãéŠéå ããã¨ããšãã¤ãˇãŧãĒéĻããæĨŊãããžãã
ã¯ããŗããŗãĒãĸãŗããŧãĒãŠãŽãšãã¤ãšãäŊŋãã¨ãéŖæŦ˛ãããããžãã
**âĄãšãã¤ãšãįãã**
ãšãã¤ãšã¯å§ããĢå°ãå
Ĩããã ãã§ãããŽåžã¯åžã
ãĢå ããžãã
**âĸįãããŽãŋããåããįãã**
įãããįãããã¨ã§ãįãŋãįãžããžãã
**âŖéļčãįãã**
éļčãįããéãĢã¯ãįĻããĒããããĢæŗ¨æããžãã
**â¤čãŽč˛ãå¤ããŖãã**
čãŽč˛ãå¤ããŖãããæŽããŽãšãã¤ãšãé įĒãĢå ããžãã
**âĨãĢãŧãã¤ãã**
æåžãĢãįĢãæĸããĻããããĢãŧãå ããžãã
**âĻãĢãŧãå ãã**
ãĢãŧãå ããéãĢã¯ãåŧąįĢã§åĒããå įąããžãã
**â§ãĢãŧãéĻ´æãã ãåŽæ**
ãĢãŧãéĻ´æãã ããåŽæã§ãã
## æåžãĢ
ãĢãŦãŧã¯ãšãã¤ãšãŽįĩãŋåããã§åŗãå¤ãããžãã
äģåį´šäģãããŦãˇããåčãĢãčĒååĨŊãŋãŽãĢãŦãŧãäŊãŖãĻãŋãĻãã ããã
åããããĢäŊãŖãĻãããšãã¤ãšãŽįĩãŋåããã§åŗãå¤ãããŽã§éĸįŊãã§ãã
äģåã¯ãšãã¤ãšãŽäŊŋãæšãžã§č§ŖčĒŦããžãããããĢãŦãŧãŽäŊãæšã¨ããĻã¯æ¯čŧįį°ĄåãĒé¨éĄã ã¨æããžãã
æ
ŖããĻããããããããŗã°ãč˛ã
å ããĻãŋãĻä¸ããã
ãĢãŦãŧã¯ãĸãŦãŗã¸ããĻãįžåŗããããéŖãšãĻæēčļŗããžãã
įãããįžåŗãããĢãŦãŧãäŊãŖãĻãŋãĻãã ããã
==========
Prompt: 6 tokens, 88.572 tokens-per-sec
Generation: 238 tokens, 85.596 tokens-per-sec
Peak memory: 2.718 GB
You can also write your code to use this model like this:
from mlx_lm import load, generate
model, tokenizer = load("mlx-community/plamo-2-1b-bf16")
prompt = "įžåŗãããĢãŦãŧãŽäŊãæšãŽãŦãˇããį´šäģããžãã"
response = generate(model, tokenizer, prompt=prompt, verbose=True)
Quantized
Base model
pfnet/plamo-2-1b