ggerganov/llama.cpp

Pull Requests Commits

naming : normalize the name of callback-related identifiers

ggerganov committed 251 days ago

Verified cc1c0171

metal : handle zero-sized allocs (#9466)

ggerganov committed 251 days ago

Verified c4965a64

flake.lock: Update (#9488)

ggerganov committed 251 days ago

Verified 90a2fff0

common : reimplement logging (#9418)

ggerganov committed 251 days ago

Verified 6262d13e

gguf-split : add basic checks (#9499)

slaren committed 251 days ago

Verified e6deac31

cmake : correct order of sycl flags (#9497)

Xarbirus committed 251 days ago

Verified 6988da94

py : add "LLaMAForCausalLM" conversion support (#9485)

csabakecskemeti committed 252 days ago

Verified 3c7989fd

readme : update tools list (#9475)

OLSecret committed 252 days ago

Verified d6b37c88

cmake : try to fix sycl+intel build (#9487)

Xarbirus committed 252 days ago

Verified 7596487b

ggml : ggml_type_name return "NONE" for invalid values (#9458)

ykhrustalev committed 253 days ago

Verified 822b6322

server: add data: [DONE] to /chat/completions stream response (#9459)

VoidIsVoid committed 253 days ago

Verified dcdcee3a

cmake : use list(APPEND ...) instead of set() + dedup linker (#9463)

ggerganov committed 253 days ago

Verified 1f4111e5

llama : make cell_id const in inp_s_mask block (#9470)

danbev committed 253 days ago

Verified befaf119

server : add loading html page while model is loading (#9468)

ngxson committed 253 days ago

Verified feff4aa8

llama : llama_perf + option to disable timings during decode (#9355)

ggerganov committed 254 days ago

Verified 0abc6a2c

feat: remove a sampler from a chain (#9445)

giladgd committed 254 days ago

Verified bd35cb0a

server : Add option to return token pieces in /tokenize endpoint (#9108)

mathijshenquet committed 254 days ago

Verified 78203641

cann: Add host buffer type for Ascend NPU (#9406)

bachelor-dou committed 254 days ago

Verified e6b7801b

llava : fix the script error in MobileVLM README (#9054)

fengerhu1 committed 254 days ago

Verified e6657443

lora : raise error if lm_head is ignored (#9103)

ngxson committed 254 days ago

Verified d4c3c10f

cmake : fix for builds without `GGML_CDEF_PUBLIC` (#9338)

Xarbirus committed 254 days ago

Verified 2a825116

ci : update HIP SDK to 24.Q3 (ROCm 6.1) (#9329)

no1wudi committed 254 days ago

Verified 4dc4f5f1

py : add Phi-1.5/Phi-2 tokenizer (#9361)

daminho committed 254 days ago

Verified c837981b

ci : bump actions/checkout to v4 (#9377)

trivikr committed 254 days ago

Verified 3c26a164

cmake : fixed the order of linking libraries for llama-quantize (#9450)

Xarbirus committed 254 days ago

Verified ff76e185

py : add special tokens in hf_converter for RWKV v6 (#9428)

MollySophia committed 254 days ago

Verified 39f852f4

riscv : modify Makefile and add a RISCV_VECT to print log info (#9442)

Tameem-10xE committed 254 days ago

Verified 2b00fa79

ggml : hide ggml_object, ggml_cgraph, ggml_hash_set (#9408)

ggerganov committed 254 days ago

Verified d6a04f87

enhance run script to be easy to change the parameters (#9448)

NeoZhangJianyu committed 255 days ago

Verified c9c8575a

cann: Fix error when running a non-exist op (#9424)

bachelor-dou committed 255 days ago

Verified df4b7945

Older