BTL-3-Compact / evidence /thinking-escape-v3-native-runtime.json
affableiq's picture
Refresh canonical Compact with thinking-escape r8 step160 v3
1ccb6ee verified
Raw
History Blame Contribute Delete
1.04 kB
{
"schema_version": 1,
"model": "BTL-3-Compact-Thinking-Escape-r8-step160-v3.gguf",
"model_sha256": "0a4d9ddee49e5aa93586a792bd4d452ea837229d49d22e54212dde87a5c9888a",
"runtime": "BTL-3-Compact-macos-arm64 packaged native runtime",
"device": "Apple Metal",
"context_size": 4096,
"default_thinking": false,
"health": {"status": "ok"},
"direct_answer": {
"prompt": "What programming language is known for snake_case and significant indentation? Answer with one word.",
"content": "Python",
"finish_reason": "stop",
"completion_tokens": 2,
"passed": true
},
"coding": {
"prompt": "Write a Python function add_one(value: int) -> int that returns value plus one. Return only code.",
"content": "```python\ndef add_one(value: int) -> int:\n return value + 1\n```",
"finish_reason": "stop",
"completion_tokens": 23,
"ast_parse": true,
"execution": {"expression": "add_one(41)", "expected": 42, "actual": 42},
"passed": true
},
"server_stopped_after_validation": true
}