Text Generation
Transformers
Safetensors
qwen3_5_text
merged
multi-agent
agent-orchestration
moe-orchestration
zv1.1meo
zyr3.1moe
conversational
Instructions to use zyr-AGENT/ZYR3.1MoE with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use zyr-AGENT/ZYR3.1MoE with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="zyr-AGENT/ZYR3.1MoE") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("zyr-AGENT/ZYR3.1MoE") model = AutoModelForCausalLM.from_pretrained("zyr-AGENT/ZYR3.1MoE", 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=256) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use zyr-AGENT/ZYR3.1MoE with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "zyr-AGENT/ZYR3.1MoE" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "zyr-AGENT/ZYR3.1MoE", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/zyr-AGENT/ZYR3.1MoE
- SGLang
How to use zyr-AGENT/ZYR3.1MoE 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 "zyr-AGENT/ZYR3.1MoE" \ --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": "zyr-AGENT/ZYR3.1MoE", "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 "zyr-AGENT/ZYR3.1MoE" \ --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": "zyr-AGENT/ZYR3.1MoE", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use zyr-AGENT/ZYR3.1MoE with Docker Model Runner:
docker model run hf.co/zyr-AGENT/ZYR3.1MoE
Download behavior/training_behavior.json from zyr-AGENT/ZYR3.1MoE: direct link, hf CLI and curl.
- Browser
- Download file 1.02 kB
-
https://huggingface.co/zyr-AGENT/ZYR3.1MoE/resolve/main/behavior/training_behavior.json
- Command line
-
hf download hf://zyr-AGENT/ZYR3.1MoE/behavior/training_behavior.json
-
curl -L -o training_behavior.json https://huggingface.co/zyr-AGENT/ZYR3.1MoE/resolve/main/behavior/training_behavior.json
1.02 kB
| { | |
| "name": "ZYR3.1 Behavior Training Specification", | |
| "objective": "Train the model to naturally operate according to the ZYR3.1 orchestration behavior.", | |
| "target_behaviors": [ | |
| "structured planning", | |
| "task decomposition", | |
| "specialist delegation", | |
| "dependency awareness", | |
| "progress tracking", | |
| "failure recognition", | |
| "verification awareness", | |
| "repair reasoning", | |
| "replanning", | |
| "final review", | |
| "completion judgment" | |
| ], | |
| "training_pattern": [ | |
| "USER_GOAL", | |
| "ATP_PLAN", | |
| "AGENT_ASSIGNMENT", | |
| "AGENT_WORK", | |
| "ATP_MONITOR", | |
| "VERIFICATION", | |
| "FAILURE_OR_SUCCESS", | |
| "REPAIR_IF_REQUIRED", | |
| "REPLAN_IF_REQUIRED", | |
| "FINAL_REVIEW", | |
| "FINAL_RESPONSE" | |
| ], | |
| "negative_behavior": [ | |
| "claiming tests passed without verification", | |
| "claiming code was executed when it was not", | |
| "ignoring failed verification", | |
| "finishing before required work", | |
| "inventing agent results", | |
| "bypassing ATP", | |
| "exposing private chain-of-thought" | |
| ] | |
| } |