19 Commits

Author	SHA1	Message	Date
turboderp	8877b99855	QCache: Skip dequant when possible outside of SWA window	2026-04-13 00:52:46 +02:00
turboderp	87830de104	Bump to v0.0.29	2026-04-12 02:27:23 +02:00
turboderp	6b8225ff41	Add bighead-attn and bighead-attn-paged kernels	2026-04-12 01:34:54 +02:00
lesj0610	69d6c6ad76	feat(gemma4): add multimodal generation support	2026-04-03 12:52:51 +09:00
lesj0610	f4bec2eb2e	feat(gemma4): add Gemma4 architecture support	2026-04-03 09:49:12 +09:00
turboderp	1592d04ffd	Tests: Fix up generator stress test	2026-03-22 18:16:33 +01:00
turboderp	15647d98d7	Sampling: Fix possible divide-by-zero in rep.penalty kernels	2026-03-22 18:02:48 +01:00
lesj0610	88062566f5	Qwen3.5: Smoke test	2026-03-02 15:49:29 +01:00
turboderp	0f2da5d6a7	GEMM: Lock MCG multiplier to 0xCBAC1FED and MUL1 to 0x83DCD12D. Make MCG the default codebook for new models.	2025-10-12 22:09:01 +02:00
turboderp	12eadfe114	Generator: Add requeue option	2025-09-22 03:34:31 +02:00
turboderp	1ff09ee3c4	Generator: Fix batching with recurrent states	2025-09-22 03:30:37 +02:00
turboderp	9be91d644d	Generator: Periodically defragment paged cache	2025-05-28 22:23:08 +02:00
turboderp	965f5b2f0a	Add cache_rotate kernel	2025-05-27 11:27:51 +02:00
turboderp	c82af98d57	New RoPE kernel with fused head norm	2025-05-25 13:33:39 +02:00
turboderp	d359bcc0d3	Add MCG 3INST and MCG 1MAD (MUL1) experimental quant modes	2025-05-21 19:15:13 +02:00
turboderp	40863d0690	Add repetition, presence and frequency penalties	2025-04-27 01:09:33 +02:00
turboderp	702ebab361	Add sampler test cases	2025-04-25 00:48:06 +02:00
turboderp	cf84811485	Add cache quantization	2025-04-22 21:52:33 +02:00
turboderp	543c4b2771	Initial commit	2025-04-06 14:42:49 +02:00