Instructions to use MTEnt/dot with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MTEnt/dot with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="MTEnt/dot") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("MTEnt/dot") model = AutoModelForCausalLM.from_pretrained("MTEnt/dot", device_map="auto") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use MTEnt/dot with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "MTEnt/dot" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "MTEnt/dot", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/MTEnt/dot
- SGLang
How to use MTEnt/dot 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 "MTEnt/dot" \ --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": "MTEnt/dot", "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 "MTEnt/dot" \ --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": "MTEnt/dot", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use MTEnt/dot with Docker Model Runner:
docker model run hf.co/MTEnt/dot
| { | |
| "release": "Dot-v0.4-Thinking", | |
| "checkpoint_step": 938, | |
| "training_tokens": 8642015, | |
| "evaluation_scope": { | |
| "task_families": 8, | |
| "relationship_to_training": "same executable generators, separately seeded and hash-disjoint records", | |
| "warning": "This is targeted curriculum evaluation, not a broad reasoning or ARC-AGI benchmark." | |
| }, | |
| "teacher_forced_4096": { | |
| "baseline_zero_gate": { | |
| "response_nll": 2.1283936659, | |
| "response_perplexity": 8.40136, | |
| "teacher_forced_token_accuracy": 0.666026, | |
| "teacher_forced_exact_response_rate": 0.0, | |
| "reasoning_nll": 2.511300, | |
| "reasoning_token_accuracy": 0.604705, | |
| "answer_nll": 0.0667919, | |
| "answer_token_accuracy": 0.9960501 | |
| }, | |
| "dot_v0_4": { | |
| "response_nll": 0.01736697, | |
| "response_perplexity": 1.017519, | |
| "teacher_forced_token_accuracy": 0.9942831, | |
| "teacher_forced_exact_response_rate": 0.8776855, | |
| "reasoning_nll": 0.0205677, | |
| "reasoning_token_accuracy": 0.9932305, | |
| "answer_nll": 0.00011697, | |
| "answer_token_accuracy": 0.9999480 | |
| }, | |
| "response_nll_change": -2.1110267 | |
| }, | |
| "free_running_256": { | |
| "thinking_envelope_valid_rate": 1.0, | |
| "nonempty_reasoning_rate": 1.0, | |
| "mean_reasoning_tokens": 23.8633, | |
| "mean_total_generated_tokens": 28.5977, | |
| "truncated_before_think_close_rate": 0.0, | |
| "final_answer_exact_match": 0.96875, | |
| "correct": 248, | |
| "examples": 256 | |
| }, | |
| "adversarial_spatial_probe_64": { | |
| "suite_relationship": "new spatial generators outside the v0.4 training curriculum", | |
| "final_answer_exact_match": 0.21875, | |
| "correct": 14, | |
| "examples": 64, | |
| "thinking_envelope_valid_rate": 0.875, | |
| "truncation_rate": 0.109375, | |
| "orientation_exact_match": 0.0, | |
| "warning": "The larger 256-case run was stopped after a 15-minute timeout caused by uncached verbose generation." | |
| } | |
| } | |