Instructions to use collision-10M/Collision-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use collision-10M/Collision-1B with Transformers:

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

pipe = pipeline("text-generation", model="collision-10M/Collision-1B", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("collision-10M/Collision-1B", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use collision-10M/Collision-1B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "collision-10M/Collision-1B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "collision-10M/Collision-1B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/collision-10M/Collision-1B

SGLang

How to use collision-10M/Collision-1B 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 "collision-10M/Collision-1B" \
    --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": "collision-10M/Collision-1B",
		"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 "collision-10M/Collision-1B" \
        --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": "collision-10M/Collision-1B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use collision-10M/Collision-1B with Docker Model Runner:
```
docker model run hf.co/collision-10M/Collision-1B
```

Collision-1B / model /blocks.py

collision-10M

Release COLLISION-10M with official in-house NLP Engine and analytics

ddc12cc verified about 1 month ago

Raw History Blame Contribute Delete

1.07 kB

import torch
import torch.nn as nn
from model.attention import CausalSelfAttention

class FeedForward(nn.Module):
    def __init__(self, d_model: int, d_ff: int, dropout: float = 0.1):
        super().__init__()
        self.net = nn.Sequential(
            nn.Linear(d_model, d_ff),
            nn.GELU(),
            nn.Linear(d_ff, d_model),
            nn.Dropout(dropout)
        )
        
    def forward(self, x: torch.Tensor) -> torch.Tensor:
        return self.net(x)

class TransformerBlock(nn.Module):
    def __init__(self, d_model: int, n_head: int, max_seq_len: int, d_ff: int, dropout: float = 0.1):
        super().__init__()
        self.ln1 = nn.LayerNorm(d_model)
        self.attn = CausalSelfAttention(d_model, n_head, max_seq_len, dropout)
        self.ln2 = nn.LayerNorm(d_model)
        self.ffn = FeedForward(d_model, d_ff, dropout)
        
    def forward(self, x: torch.Tensor) -> torch.Tensor:
        # Pre-LN structure with residual connections
        x = x + self.attn(self.ln1(x))
        x = x + self.ffn(self.ln2(x))
        return x

	import torch
	import torch.nn as nn
	from model.attention import CausalSelfAttention

	class FeedForward(nn.Module):
	def __init__(self, d_model: int, d_ff: int, dropout: float = 0.1):
	super().__init__()
	self.net = nn.Sequential(
	nn.Linear(d_model, d_ff),
	nn.GELU(),
	nn.Linear(d_ff, d_model),
	nn.Dropout(dropout)
	)

	def forward(self, x: torch.Tensor) -> torch.Tensor:
	return self.net(x)

	class TransformerBlock(nn.Module):
	def __init__(self, d_model: int, n_head: int, max_seq_len: int, d_ff: int, dropout: float = 0.1):
	super().__init__()
	self.ln1 = nn.LayerNorm(d_model)
	self.attn = CausalSelfAttention(d_model, n_head, max_seq_len, dropout)
	self.ln2 = nn.LayerNorm(d_model)
	self.ffn = FeedForward(d_model, d_ff, dropout)

	def forward(self, x: torch.Tensor) -> torch.Tensor:
	# Pre-LN structure with residual connections
	x = x + self.attn(self.ln1(x))
	x = x + self.ffn(self.ln2(x))
	return x