Text Generation
MLX
Safetensors
zaya
mixture-of-experts
hybrid-attention
cca-attention
apple-silicon
reasoning
tool-use
quantized
jang
jangtq
jangtq-k
mixed-precision
mxtq
jangtq-prestack
osaurus
conversational
Instructions to use OsaurusAI/ZAYA1-8B-JANGTQ_K with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use OsaurusAI/ZAYA1-8B-JANGTQ_K with MLX:
# Make sure mlx-lm is installed # pip install --upgrade mlx-lm # Generate text with mlx-lm from mlx_lm import load, generate model, tokenizer = load("OsaurusAI/ZAYA1-8B-JANGTQ_K") prompt = "Write a story about Einstein" messages = [{"role": "user", "content": prompt}] prompt = tokenizer.apply_chat_template( messages, add_generation_prompt=True ) text = generate(model, tokenizer, prompt=prompt, verbose=True) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
- Pi new
How to use OsaurusAI/ZAYA1-8B-JANGTQ_K with Pi:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "OsaurusAI/ZAYA1-8B-JANGTQ_K"
Configure the model in Pi
# Install Pi: npm install -g @mariozechner/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "mlx-lm": { "baseUrl": "http://localhost:8080/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "OsaurusAI/ZAYA1-8B-JANGTQ_K" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Hermes Agent new
How to use OsaurusAI/ZAYA1-8B-JANGTQ_K with Hermes Agent:
Start the MLX server
# Install MLX LM: uv tool install mlx-lm # Start a local OpenAI-compatible server: mlx_lm.server --model "OsaurusAI/ZAYA1-8B-JANGTQ_K"
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:8080/v1 hermes config set model.default OsaurusAI/ZAYA1-8B-JANGTQ_K
Run Hermes
hermes
- MLX LM
How to use OsaurusAI/ZAYA1-8B-JANGTQ_K with MLX LM:
Generate or start a chat session
# Install MLX LM uv tool install mlx-lm # Interactive chat REPL mlx_lm.chat --model "OsaurusAI/ZAYA1-8B-JANGTQ_K"
Run an OpenAI-compatible server
# Install MLX LM uv tool install mlx-lm # Start the server mlx_lm.server --model "OsaurusAI/ZAYA1-8B-JANGTQ_K" # Calling the OpenAI-compatible server with curl curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "OsaurusAI/ZAYA1-8B-JANGTQ_K", "messages": [ {"role": "user", "content": "Hello"} ] }'
| { | |
| "activation_func": "swiglu", | |
| "activation_func_fp8_input_store": false, | |
| "add_bias_linear": false, | |
| "architectures": [ | |
| "ZayaForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bias_activation_fusion": true, | |
| "bos_token_id": 2, | |
| "cca": true, | |
| "cca_num_q_heads": 8, | |
| "dtype": "bfloat16", | |
| "eos_token_id": 106, | |
| "ffn_hidden_size": 4096, | |
| "gated_linear_unit": true, | |
| "hidden_size": 2048, | |
| "kv_channels": 128, | |
| "lm_head_bias": false, | |
| "mamba_cache_dtype": "float32", | |
| "max_position_embeddings": 131072, | |
| "model_type": "zaya", | |
| "moe_router_topk": 1, | |
| "norm_epsilon": 1e-05, | |
| "normalization": "RMSNorm", | |
| "num_attention_heads": 16, | |
| "num_experts": 16, | |
| "num_hidden_layers": 80, | |
| "num_key_value_heads": 2, | |
| "num_query_groups": 2, | |
| "pad_token_id": 0, | |
| "partial_rotary_factor": 0.5, | |
| "residual_in_fp32": true, | |
| "rope_scaling": false, | |
| "rope_theta": 5000000, | |
| "scale_residual_merge": true, | |
| "sliding_window": null, | |
| "transformers_version": "4.57.1", | |
| "use_cache": true, | |
| "vocab_size": 262272, | |
| "zaya_mlp_expansion": 256, | |
| "zaya_use_eda": true, | |
| "zaya_use_mod": true, | |
| "capabilities": { | |
| "reasoning_parser": "qwen3", | |
| "tool_parser": "zaya_xml", | |
| "think_in_template": false, | |
| "supports_tools": true, | |
| "supports_thinking": true, | |
| "family": "zaya", | |
| "modality": "text", | |
| "cache_type": "hybrid" | |
| }, | |
| "weight_format": "mxtq", | |
| "zaya_expert_layout": "split_switch_mlp", | |
| "tie_word_embeddings": true, | |
| "mxtq_bits": { | |
| "routed_expert": { | |
| "gate_proj": 2, | |
| "up_proj": 2, | |
| "down_proj": 4 | |
| }, | |
| "attention": 8, | |
| "router": 16, | |
| "embed_tokens": 8, | |
| "lm_head": 8, | |
| "cca_conv": 16, | |
| "norms_residual": 16 | |
| }, | |
| "mxtq_seed": 42, | |
| "quantization": { | |
| "bits": 8, | |
| "group_size": 32, | |
| "mode": "affine", | |
| "routed_expert_bits": { | |
| "gate_proj": 2, | |
| "up_proj": 2, | |
| "down_proj": 4 | |
| }, | |
| "mxtq_bits": { | |
| "routed_expert": { | |
| "gate_proj": 2, | |
| "up_proj": 2, | |
| "down_proj": 4 | |
| }, | |
| "attention": 8, | |
| "router": 16, | |
| "embed_tokens": 8, | |
| "lm_head": 8, | |
| "cca_conv": 16, | |
| "norms_residual": 16 | |
| }, | |
| "expert_layout": "split_switch_mlp" | |
| } | |
| } |