ochafik/llama.cpp

Pull Requests Commits
readme : add programmable prompt engine language CLI (#9599)
snowyu committed 2 years ago
Verified 1d48e98e
Update json-schema-to-grammar.cpp
ochafik committed 2 years ago
071f2a7b
json: make SchemaConverter.add_rule public
ochafik committed 2 years ago
a50a24c3
flake.lock: Update (#9586)
ggerganov committed 2 years ago
Verified f3979df7
ggml : AVX512 gemm for Q4_0_8_8 (#9532)
Srihari-mcw committed 2 years ago
Verified 1e7b9299
perplexity : remove extra new lines after chunks (#9596)
ggerganov committed 2 years ago
Verified 37f8c7b4
metal : use F32 prec for K*Q in vec FA (#9595)
ggerganov committed 2 years ago
Verified bf9c1013
Revert "[SYCL] fallback mmvq (#9088)" (#9579)
Akarshan Biswas committed 2 years ago
Verified e62e9789
update jinja test cases
ochafik committed 2 years ago
22b33768
jinja: support selectattr, last, map, length on str
ochafik committed 2 years ago
31139bbc
fix crash
ochafik committed 2 years ago
fb179627
musa: enable building fat binaries, enable unified memory, and disable Flash Attention on QY1 (MTT S80) (#9526)
yeahdongcn committed 2 years ago
Verified c35e586e
jinja: implement str.strip()
ochafik committed 2 years ago
80cf8a76
jinja: pass context as shared ptr
ochafik committed 2 years ago
b85e7b60
Update utils.hpp
ochafik committed 2 years ago
10b15906
jinja: fix precedence of not vs. is operators
ochafik committed 2 years ago
eea07c57
Update utils.hpp
ochafik committed 2 years ago
5a7e55bd
wire jinja through normal message paths
ochafik committed 2 years ago
bfd81c78
Fix merge error in #9454 (#9589)
MollySophia committed 2 years ago
Verified 912c331d
CUDA: enable Gemma FA for HIP/Pascal (#9581)
JohannesGaessler committed 2 years ago
Verified a5b57b08
llama: remove redundant loop when constructing ubatch (#9574)
shankarg87 committed 2 years ago
Verified ecd5d6b6
RWKV v6: RWKV_WKV op CUDA implementation (#9454)
MollySophia committed 2 years ago
Verified 2a63caaa
minja goldens
ochafik committed 2 years ago
8186f35c
(DRAFT) use minja in test-chat-template
ochafik committed 2 years ago
70a97f1e
ggml-alloc : fix list of allocated tensors with GGML_ALLOCATOR_DEBUG (#9573)
slaren committed 2 years ago
Verified d09770ca
Update CUDA graph on scale change plus clear nodes/params (#9550)
agray3 committed 2 years ago
Verified 41f47787
CI: Provide prebuilt windows binary for hip (#9467)
no1wudi committed 2 years ago
Verified e948a7da
quantize : improve type name parsing (#9570)
slaren committed 2 years ago
Verified 63351143
ggml : fix builds (#0)
ggerganov committed 2 years ago
d13edb17
ggml : fix trailing whitespace (#0)
ggerganov committed 2 years ago
27609c49
Terms Of Use Privacy Policy Refund Policy Impressum