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

File size: 1,347 Bytes

{
  "name": "COLLISION-1B",
  "version": "1.0.0",
  "parameters": 999376128,
  "architecture": "Decoder-only Transformer",
  "layers": 24,
  "hidden_size": 2048,
  "attention_heads": 16,
  "d_ff": 5376,
  "context_length": 1024,
  "vocab_size": 32000,
  "training_tokens": 10000000,
  "dataset": "collision_dataset_v5_expanded",
  "tokenizer": "Custom BPE Tokenizer",
  "checkpoint_sha256": "bdd986e2a4964a6a204224dbd973625abe192cd4f6e23dceb79e273a29b19c88",
  "nlp_engine": {
    "module": "collision.nlp",
    "version": "1.0.0",
    "capabilities": [
      "conversational_dialogue",
      "textrank_keyphrase_extraction",
      "topic_classification_10_domains",
      "tone_and_formality_scoring",
      "grammar_and_spell_proofreading",
      "readability_indices_flesch_fog",
      "reading_comprehension_context_qa",
      "style_transformations",
      "semantic_text_similarity",
      "deterministic_math_geometry_statistics",
      "grounded_answer_synthesis"
    ]
  },
  "cognitive_brain": {
    "module": "collision.brain",
    "version": "2.0.0",
    "capabilities": [
      "dual_process_system_1_2",
      "graph_of_thoughts_dag",
      "hegelian_dialectic_synthesis",
      "global_workspace_theory",
      "synaptic_working_memory_ltp",
      "cross_domain_knowledge_lattice",
      "information_entropy_profiling"
    ]
  }
}