Instructions to use AtomixLabs/Photon-2.0-1M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AtomixLabs/Photon-2.0-1M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="AtomixLabs/Photon-2.0-1M")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("AtomixLabs/Photon-2.0-1M") model = AutoModelForCausalLM.from_pretrained("AtomixLabs/Photon-2.0-1M", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use AtomixLabs/Photon-2.0-1M with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "AtomixLabs/Photon-2.0-1M" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AtomixLabs/Photon-2.0-1M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/AtomixLabs/Photon-2.0-1M
- SGLang
How to use AtomixLabs/Photon-2.0-1M 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 "AtomixLabs/Photon-2.0-1M" \ --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": "AtomixLabs/Photon-2.0-1M", "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 "AtomixLabs/Photon-2.0-1M" \ --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": "AtomixLabs/Photon-2.0-1M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use AtomixLabs/Photon-2.0-1M with Docker Model Runner:
docker model run hf.co/AtomixLabs/Photon-2.0-1M
ouuhhh shi a new worthless model
new dogshit model that cant do anything and is uselessπππππππππππππ
new dogshit model that cant do anything and is uselessπππππππππππππ
It's a 1M parameter experimental research model! The fact that you expected general chat capabilities from a model smaller than an image file says far more about your reading comprehension than our architecture.
new dogshit model that cant do anything and is uselessπππππππππππππ
It's a 1M parameter experimental research model! The fact that you expected general chat capabilities from a model smaller than an image file says far more about your reading comprehension than our architecture.
"research" if u cant do anything impressive like general chat capabilities on 500k params js dont do it we already have enough slm research(useless) models that do nothing for humanity
new dogshit model that cant do anything and is uselessπππππππππππππ
It's a 1M parameter experimental research model! The fact that you expected general chat capabilities from a model smaller than an image file says far more about your reading comprehension than our architecture.
"research" if u cant do anything impressive like general chat capabilities on 500k params js dont do it we already have enough slm research(useless) models that do nothing for humanity
Expecting conversational AI on 500k parameters violates basic information theory. If you figured out how to fit human language and reasoning into half a million floating-point numbers, publish the paper because you just won a Turing Award. Until then, let the researchers work.
new dogshit model that cant do anything and is uselessπππππππππππππ
It's a 1M parameter experimental research model! The fact that you expected general chat capabilities from a model smaller than an image file says far more about your reading comprehension than our architecture.
"research" if u cant do anything impressive like general chat capabilities on 500k params js dont do it we already have enough slm research(useless) models that do nothing for humanity
Expecting conversational AI on 500k parameters violates basic information theory. If you figured out how to fit human language and reasoning into half a million floating-point numbers, publish the paper because you just won a Turing Award. Until then, let the researchers work.
yea yea bro information theory so stop doing research if u cant understand the basic concept of research and experimentationπβοΈ
new dogshit model that cant do anything and is uselessπππππππππππππ
It's a 1M parameter experimental research model! The fact that you expected general chat capabilities from a model smaller than an image file says far more about your reading comprehension than our architecture.
"research" if u cant do anything impressive like general chat capabilities on 500k params js dont do it we already have enough slm research(useless) models that do nothing for humanity
Expecting conversational AI on 500k parameters violates basic information theory. If you figured out how to fit human language and reasoning into half a million floating-point numbers, publish the paper because you just won a Turing Award. Until then, let the researchers work.
yea yea bro information theory so stop doing research if u cant understand the basic concept of research and experimentationπβοΈ
From calling research useless to trying to explain research to researchers in two comments. The mental gymnastics are impressive. Take care!