{"data": [{"schema_version": "2.4", "id": "nova26/ternary-bonsai-2-2b", "hugging_face_id": "prism-ml/Ternary-Bonsai-2-2B-gguf", "name": "Ternary-Bonsai-2-2B (Q2_0)", "description": "PrismML Ternary-Bonsai-2-2B (Qwen3.5 hybrid, 1.9B params) in its Q2_0 ternary pack, served exactly as named on pinned upstream llama.cpp from a fleet-owned machine in the US. Built for short, decision-shaped work: routing, classification, extraction. Prompts are never logged (metadata only) and never used for training.", "input_modalities": [{"type": "text", "supported_inputs": {"max_context_length": {"value": 131072, "unit": "token"}}, "pricing": [{"type": "prompt", "unit": "token", "cost_usd": "0.0000008"}], "capacity": [{"type": "prompt", "unit": "token", "per": "minute", "value": 60000}]}], "output_modalities": [{"type": "text", "streaming": true, "max_length": {"value": 8192, "unit": "token"}, "supported_parameters": {"temperature": {"type": "range", "min": 0, "max": 2}, "top_p": {"type": "range", "min": 0, "max": 1}, "top_k": {"type": "integer", "min": 0, "max": 1000}, "seed": {"type": "integer", "min": 0, "max": 4294967295}, "max_tokens": {"type": "integer", "min": 1, "max": 8192, "unit": "token"}, "stop": {"type": "array", "items": {"type": "unknown"}, "max_items": 4}}, "pricing": [{"type": "completion", "unit": "token", "cost_usd": "0.000001"}]}], "datacenters": [{"country_code": "US", "region": "us-east-1"}], "deployment_region": "US"}]}