Georgi Gerganov 2fbde785bc kv-cache : optimize KQ mask construction (#18842) ...

* kv-cache : optimize KQ mask construction

* cont : add explanation + improve

* cont : fix

2026-01-17 15:42:42 +02:00

..

models

model : add EXAONE MoE (#18543)

2026-01-13 23:28:38 +01:00

CMakeLists.txt

model : add EXAONE MoE (#18543)

2026-01-13 23:28:38 +01:00

llama-adapter.cpp

lora: make sure model keep track of associated adapters (#18490)

2026-01-15 10:24:28 +01:00

llama-adapter.h

lora: make sure model keep track of associated adapters (#18490)

2026-01-15 10:24:28 +01:00

llama-arch.cpp

model : add EXAONE MoE (#18543)

2026-01-13 23:28:38 +01:00

llama-arch.h

model : add EXAONE MoE (#18543)

2026-01-13 23:28:38 +01:00

llama-batch.cpp

batch : fix sequence id ownership (#17915)

2025-12-11 14:29:47 +02:00

llama-batch.h

batch : fix sequence id ownership (#17915)

2025-12-11 14:29:47 +02:00

llama-chat.cpp

model : add EXAONE MoE (#18543)

2026-01-13 23:28:38 +01:00

llama-chat.h

model : add EXAONE MoE (#18543)

2026-01-13 23:28:38 +01:00

llama-context.cpp

context : do not reserve scheduler for warmups (#18867)

2026-01-15 19:35:57 +02:00

llama-context.h

context : reserve new scheduler when graph topology changes (#18547)

2026-01-15 16:39:17 +02:00

llama-cparams.cpp

cparams : rename LLAMA_MAX_PARALLEL_SEQUENCES to LLAMA_MAX_SEQ (#14188)

2025-06-15 10:08:58 +03:00

llama-cparams.h

context : reserve new scheduler when graph topology changes (#18547)

2026-01-15 16:39:17 +02:00

llama-grammar.cpp

common/grammar : replace problematic backtracking regex [\s\S]* (#18342)

2026-01-03 16:02:43 -06:00

llama-grammar.h

common/grammar : replace problematic backtracking regex [\s\S]* (#18342)

2026-01-03 16:02:43 -06:00

llama-graph.cpp

graph : clean up t5 input builders (#18795)

2026-01-13 09:43:51 +01:00

llama-graph.h

sampling : add support for backend sampling (#17004)

2026-01-04 22:22:16 +02:00

llama-hparams.cpp

kv-cache : optimize KQ mask construction (#18842)

2026-01-17 15:42:42 +02:00

llama-hparams.h

kv-cache : optimize KQ mask construction (#18842)

2026-01-17 15:42:42 +02:00

llama-impl.cpp

llama: automatically set parameters not set by the user in such a way that maximizes GPU utilization (#16653)

2025-12-15 09:24:59 +01:00

llama-impl.h

ggml, llama : use defaulted constructors/destructors (#17649)

2025-12-03 07:12:18 +01:00

llama-io.cpp

llama : refactor llama_context, llama_kv_cache, llm_build_context (#12181)

2025-03-13 12:35:44 +02:00

llama-io.h

llama : refactor llama_context, llama_kv_cache, llm_build_context (#12181)

2025-03-13 12:35:44 +02:00

llama-kv-cache-iswa.cpp

kv-cache : pad the cache size to 256 for performance (#17046)

2025-11-07 20:03:25 +02:00

llama-kv-cache-iswa.h

llama: print memory breakdown on exit (#15860)

2025-09-24 16:53:48 +02:00

llama-kv-cache.cpp

kv-cache : optimize KQ mask construction (#18842)

2026-01-17 15:42:42 +02:00

llama-kv-cache.h

kv-cache : optimize KQ mask construction (#18842)

2026-01-17 15:42:42 +02:00

llama-kv-cells.h

llama: store mrope data in KV cell (#16825)

2025-10-29 18:09:18 +01:00

llama-memory-hybrid.cpp

graph : reuse SSM graphs (#16490)

2025-12-16 09:36:21 +02:00

llama-memory-hybrid.h

llama: print memory breakdown on exit (#15860)

2025-09-24 16:53:48 +02:00

llama-memory-recurrent.cpp

memory: Hybrid context shift (#17009)

2025-11-10 17:14:23 +02:00

llama-memory-recurrent.h

llama: consistent ctx <-> buf order for KV cache (#16746)

2025-10-28 11:23:54 +01:00

llama-memory.cpp

memory : correctly handle failure in apply() (#14438)

2025-06-30 18:03:03 +03:00

llama-memory.h

llama: print memory breakdown on exit (#15860)

2025-09-24 16:53:48 +02:00

llama-mmap.cpp

mmap: add Haiku support by skipping RLIMIT_MEMLOCK check (#18819)

2026-01-14 09:11:05 +02:00

llama-mmap.h

llama : add use_direct_io flag for model loading (#18166)

2026-01-08 08:35:30 +02:00

llama-model-loader.cpp

model-loader : support bool array sliding window pattern (#18850)

2026-01-15 10:12:46 +01:00

llama-model-loader.h

llama : add use_direct_io flag for model loading (#18166)

2026-01-08 08:35:30 +02:00

llama-model-saver.cpp

model : add LFM2-ColBert-350M (#18607)

2026-01-05 19:52:56 +01:00

llama-model-saver.h

llama/ggml: add LLM training support (#10544)

2025-05-12 14:44:49 +02:00

llama-model.cpp

lora: make sure model keep track of associated adapters (#18490)

2026-01-15 10:24:28 +01:00

llama-model.h

lora: make sure model keep track of associated adapters (#18490)

2026-01-15 10:24:28 +01:00

llama-quant.cpp

llama : add use_direct_io flag for model loading (#18166)

2026-01-08 08:35:30 +02:00

llama-quant.h

llama : refactor src/llama.cpp (#10902)

2025-01-03 10:18:53 +02:00

llama-sampling.cpp

llama : add adaptive-p sampler (#17927)

2026-01-15 19:16:29 +02:00

llama-sampling.h

sampling : add support for backend sampling (#17004)

2026-01-04 22:22:16 +02:00

llama-vocab.cpp

llama : print_info alignment fix (#18708)

2026-01-14 00:05:11 +01:00

llama-vocab.h

model : add EXAONE MoE (#18543)

2026-01-13 23:28:38 +01:00

llama.cpp

llama: use host memory if device reports 0 memory (#18587)

2026-01-09 05:34:56 +08:00

unicode-data.cpp

server : better security control for public deployments (#9776)

2024-10-08 13:27:04 +02:00

unicode-data.h

llama : reduce compile time and binary size (#9712)

2024-10-02 15:49:55 +02:00

unicode.cpp

model: support youtu-vl model (#18479)

2026-01-01 19:25:54 +01:00

unicode.h

devops: add s390x & ppc64le CI (#15925)

2025-09-27 02:03:33 +08:00