mirror of
https://github.com/ikawrakow/ik_llama.cpp.git
synced 2026-05-13 09:26:00 +00:00
* qwen-mtp: add dense mtp for one draft * add support for smaller qwen mtp commit * qwen-mtp: fix graph for qwen dense variants * Squashed commit of the following: commit a92a154b38c7fddc84460f8852c900f8d6ce907e Author: SamuelOliveirads <samueloliveira32df@gmail.com> Date: Mon Apr 20 13:30:21 2026 -0300 recurrent model: refactor api commitdfac8f19f6Author: SamuelOliveirads <samueloliveira32df@gmail.com> Date: Mon Apr 20 12:22:29 2026 -0300 recurrent model: implement recurrent kernel checkpoint commit9c44b117f9Author: SamuelOliveirads <samueloliveira32df@gmail.com> Date: Sat Apr 18 11:52:39 2026 -0300 speculative: fix sampler for checkpoints commite7006393bcAuthor: SamuelOliveirads <samueloliveira32df@gmail.com> Date: Fri Apr 17 14:08:25 2026 -0300 server: refactor checkpoint state logic commit57eabf04dfMerge:dc4797b764234e3cAuthor: SamuelOliveirads <samueloliveira32df@gmail.com> Date: Fri Apr 17 13:53:41 2026 -0300 Merge branch 'main' into fix/hybrid-cache-speculative commitdc4797b723Author: SamuelOliveirads <samueloliveira32df@gmail.com> Date: Fri Apr 17 13:12:40 2026 -0300 reset ngram mod state for rejected tokens commit8ff2d943a3Author: SamuelOliveirads <samueloliveira32df@gmail.com> Date: Fri Apr 17 13:08:04 2026 -0300 server: snapshot recurrent state in tensor commitd93dfb5e6bAuthor: SamuelOliveirads <samueloliveira32df@gmail.com> Date: Thu Apr 16 22:36:37 2026 -0300 fix: save/restore sampler state during speculative checkpoint When speculative decoding rejects draft tokens and restores the recurrent state checkpoint, the sampler (RNG, grammar, prev tokens) must also be restored to maintain consistency. Without this, the sampler state reflects the rejected draft tokens, leading to potential divergence. Uses common_sampler_clone() to snapshot the sampler before the speculative batch decode, and restores it on rejection. commitd670cf85cdAuthor: SamuelOliveirads <samueloliveira32df@gmail.com> Date: Thu Apr 16 21:53:52 2026 -0300 server: spec checkpoints for recurrent models * server: fix leak context between requests * qwen3: allow mtp to run with split graph * qwen3 mtp: selects rows before the ffn
182 KiB
182 KiB