32 17 minutes ago

65B dense model built for agentic coding, with up to 1M tokens of context.

tools thinking 65b
29ffb44a6b91 ยท 190B
{
"_from_model_config": true,
"bos_token_id": 100257,
"eos_token_id": [
100265,
100257
],
"pad_token_id": 100277,
"transformers_version": "5.12.1",
"use_cache": true
}