a reasoning model finetuned from Qwen2.5-32B-Instruct on just 1,000 examples. It matches o1-preview & exhibits test-time scaling via budget forcing. available in [F16, q8_0, q6_K, q4_K_S]
tools
71 Pulls Updated 8 weeks ago
b097aba8c0ca · 113B
{
"num_ctx": 131072,
"repeat_penalty": 1.1,
"stop": [
"<|im_end|>"
],
"temperature": 0.1,
"top_k": 20,
"top_p": 0.7
}