| .. |
|
CMakeLists.txt
|
edc4a29eff
memory : Hybrid recurrent cache (#13979)
|
vor 7 Monaten |
|
llama-adapter.cpp
|
27ebfcacba
llama : do not crash if there is no CPU backend (#13395)
|
vor 8 Monaten |
|
llama-adapter.h
|
e0dbec0bc6
llama : refactor llama_context, llama_kv_cache, llm_build_context (#12181)
|
vor 10 Monaten |
|
llama-arch.cpp
|
e0cb5c5cb8
model : add EXAONE 4.0 support (#14630)
|
vor 6 Monaten |
|
llama-arch.h
|
e0cb5c5cb8
model : add EXAONE 4.0 support (#14630)
|
vor 6 Monaten |
|
llama-batch.cpp
|
01612b7409
llama : reuse compute graphs (#14482)
|
vor 6 Monaten |
|
llama-batch.h
|
01612b7409
llama : reuse compute graphs (#14482)
|
vor 6 Monaten |
|
llama-chat.cpp
|
e0cb5c5cb8
model : add EXAONE 4.0 support (#14630)
|
vor 6 Monaten |
|
llama-chat.h
|
e0cb5c5cb8
model : add EXAONE 4.0 support (#14630)
|
vor 6 Monaten |
|
llama-context.cpp
|
d498af3d5a
graph : avoid huge warm-up graphs for MoE models (#14753)
|
vor 6 Monaten |
|
llama-context.h
|
8f974bc1e9
graph : refactor context to not pass gf explicitly (#14629)
|
vor 6 Monaten |
|
llama-cparams.cpp
|
c311ac664d
cparams : rename LLAMA_MAX_PARALLEL_SEQUENCES to LLAMA_MAX_SEQ (#14188)
|
vor 7 Monaten |
|
llama-cparams.h
|
225e7a1438
llama : add high-throughput mode (#14363)
|
vor 6 Monaten |
|
llama-grammar.cpp
|
f5cd27b71d
`server`: streaming of tool calls and thoughts when `--jinja` is on (#12379)
|
vor 7 Monaten |
|
llama-grammar.h
|
669912d9a5
`tool-call`: fix Qwen 2.5 Coder support, add micro benchmarks, support trigger patterns for lazy grammars (#12034)
|
vor 10 Monaten |
|
llama-graph.cpp
|
bf9087f59a
metal : fuse add, mul + add tests (#14596)
|
vor 6 Monaten |
|
llama-graph.h
|
8f974bc1e9
graph : refactor context to not pass gf explicitly (#14629)
|
vor 6 Monaten |
|
llama-hparams.cpp
|
225e7a1438
llama : add high-throughput mode (#14363)
|
vor 6 Monaten |
|
llama-hparams.h
|
225e7a1438
llama : add high-throughput mode (#14363)
|
vor 6 Monaten |
|
llama-impl.cpp
|
53ff6b9b9f
GGUF: C++ refactor, backend support, misc fixes (#11030)
|
vor 1 Jahr |
|
llama-impl.h
|
fef0cbeadf
cleanup: fix compile warnings associated with gnu_printf (#11811)
|
vor 11 Monaten |
|
llama-io.cpp
|
e0dbec0bc6
llama : refactor llama_context, llama_kv_cache, llm_build_context (#12181)
|
vor 10 Monaten |
|
llama-io.h
|
e0dbec0bc6
llama : refactor llama_context, llama_kv_cache, llm_build_context (#12181)
|
vor 10 Monaten |
|
llama-kv-cache-unified-iswa.cpp
|
225e7a1438
llama : add high-throughput mode (#14363)
|
vor 6 Monaten |
|
llama-kv-cache-unified-iswa.h
|
225e7a1438
llama : add high-throughput mode (#14363)
|
vor 6 Monaten |
|
llama-kv-cache-unified.cpp
|
d6fb3f6b49
kv-cache : fix k-shift for multiple streams (#14742)
|
vor 6 Monaten |
|
llama-kv-cache-unified.h
|
01612b7409
llama : reuse compute graphs (#14482)
|
vor 6 Monaten |
|
llama-kv-cells.h
|
a70c8a0c4b
kv-cache : use ggml_set_rows (#14285)
|
vor 6 Monaten |
|
llama-memory-hybrid.cpp
|
496957e1cb
llama : fix parameter order for hybrid memory initialization (#14725)
|
vor 6 Monaten |
|
llama-memory-hybrid.h
|
a70c8a0c4b
kv-cache : use ggml_set_rows (#14285)
|
vor 6 Monaten |
|
llama-memory-recurrent.cpp
|
01612b7409
llama : reuse compute graphs (#14482)
|
vor 6 Monaten |
|
llama-memory-recurrent.h
|
692e3cdd0a
memory : rename interface to llama_memory_context_i (#14296)
|
vor 7 Monaten |
|
llama-memory.cpp
|
745f11fed0
memory : correctly handle failure in apply() (#14438)
|
vor 6 Monaten |
|
llama-memory.h
|
745f11fed0
memory : correctly handle failure in apply() (#14438)
|
vor 6 Monaten |
|
llama-mmap.cpp
|
3a077146a4
llama : allow using mmap without PrefetchVirtualMemory, apply GGML_WIN_VER to llama.cpp sources (#14013)
|
vor 7 Monaten |
|
llama-mmap.h
|
19b392d58d
llama-mmap: fix missing include (#11796)
|
vor 11 Monaten |
|
llama-model-loader.cpp
|
d17a809ef0
llama : support multiple classifier outputs and labels (#13940)
|
vor 7 Monaten |
|
llama-model-loader.h
|
e0e912f49b
llama : add option to override model tensor buffers (#11397)
|
vor 9 Monaten |
|
llama-model-saver.cpp
|
88fc854b4b
llama : improve sep token handling (#14272)
|
vor 7 Monaten |
|
llama-model-saver.h
|
10d2af0eaa
llama/ggml: add LLM training support (#10544)
|
vor 8 Monaten |
|
llama-model.cpp
|
eacdeb5bfc
model : fix build after merge conflict (#14754)
|
vor 6 Monaten |
|
llama-model.h
|
cb887f1bc1
model: add Ernie 4.5 MoE support (#14658)
|
vor 6 Monaten |
|
llama-quant.cpp
|
982e347255
quantize : fix minor logic flaw in --tensor-type (#14572)
|
vor 6 Monaten |
|
llama-quant.h
|
f66f582927
llama : refactor `src/llama.cpp` (#10902)
|
vor 1 Jahr |
|
llama-sampling.cpp
|
f9cd68398b
sampling : make sure samplers return at least 1 token (#13822)
|
vor 7 Monaten |
|
llama-sampling.h
|
afa8a9ec9b
llama : add `llama_vocab`, functions -> methods, naming (#11110)
|
vor 1 Jahr |
|
llama-vocab.cpp
|
e0cb5c5cb8
model : add EXAONE 4.0 support (#14630)
|
vor 6 Monaten |
|
llama-vocab.h
|
ab14019821
Support diffusion models: Add Dream 7B (#14644)
|
vor 6 Monaten |
|
llama.cpp
|
6adc3c3ebc
llama : add thread safety test (#14035)
|
vor 7 Monaten |
|
unicode-data.cpp
|
458367a906
server : better security control for public deployments (#9776)
|
vor 1 Jahr |
|
unicode-data.h
|
a39ab216aa
llama : reduce compile time and binary size (#9712)
|
vor 1 Jahr |
|
unicode.cpp
|
4a4f426944
model : add Kimi-K2 support (#14654)
|
vor 6 Monaten |
|
unicode.h
|
4a4f426944
model : add Kimi-K2 support (#14654)
|
vor 6 Monaten |
Georgi Gerganov