Text Generation
Transformers
PyTorch
English
crystalcoder
llm
code
custom_code
Eval Results (legacy)
Instructions to use IFM/CrystalChat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use IFM/CrystalChat with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="IFM/CrystalChat", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("IFM/CrystalChat", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use IFM/CrystalChat with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "IFM/CrystalChat" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "IFM/CrystalChat", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/IFM/CrystalChat
- SGLang
How to use IFM/CrystalChat 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 "IFM/CrystalChat" \ --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": "IFM/CrystalChat", "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 "IFM/CrystalChat" \ --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": "IFM/CrystalChat", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use IFM/CrystalChat with Docker Model Runner:
docker model run hf.co/IFM/CrystalChat
File size: 1,111 Bytes
bd4b02d 87207c5 bd4b02d 87207c5 bd4b02d 87207c5 bd4b02d | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 | {
"_name_or_path": "LLM360/CrystalChat",
"activation_function": "swiglu",
"architectures": [
"CrystalCoderLMHeadModel"
],
"attn_pdrop": 0.0,
"auto_map": {
"AutoConfig": "LLM360/CrystalChat--configuration_crystalcoder.CrystalCoderConfig",
"AutoModel": "LLM360/CrystalChat--modeling_crystalcoder.CrystalCoderModel",
"AutoModelForCausalLM": "LLM360/CrystalChat--modeling_crystalcoder.CrystalCoderLMHeadModel"
},
"bos_token_id": 1,
"embd_pdrop": 0.0,
"eos_token_id": 2,
"initializer_range": 0.02,
"layer_norm_epsilon": 1e-05,
"model_type": "crystalcoder",
"mup_embeddings_scale": 14.6,
"mup_output_alpha": 2.22,
"mup_scale_qk_dot_by_d": true,
"mup_width_scale": 0.0625,
"n_embd": 4096,
"n_head": 32,
"n_inner": 10922,
"n_layer": 32,
"n_positions": 2048,
"position_embedding_type": "rotary",
"reorder_and_upcast_attn": false,
"resid_pdrop": 0.0,
"rotary_dim": 32,
"scale_attn_by_inverse_layer_idx": false,
"scale_attn_weights": true,
"torch_dtype": "bfloat16",
"transformers_version": "4.34.0",
"use_cache": true,
"vocab_size": 32032
}
|