Tags
Tags give the ability to mark specific points in history as being important
master-0c56923
0c569234
·
examples : add save_load_state example (#1150)
·
Apr 24, 2023
master-957c8ae
957c8ae2
·
llama : increase scratch buffer size for 65B (ref #1152)
·
Apr 24, 2023
master-9b0a4d4
9b0a4d42
·
examples/main README improvements and some light refactoring (#1131)
·
Apr 24, 2023
master-2ec8342
2ec83428
·
Fix build for gcc 8 and test in CI (#1154)
·
Apr 24, 2023
master-e4cf982
e4cf982e
·
Fix cuda compilation (#1128)
·
Apr 24, 2023
master-c4fe84f
c4fe84fb
·
llama : refactor get / set state + remove redundant kv cache API (#1143)
·
Apr 24, 2023
master-ec9cdb6
ec9cdb67
·
ggml : do not print perf ops that have not been used at all
·
Apr 23, 2023
master-e4422e2
e4422e29
·
ggml : better PERF prints + support "LLAMA_PERF=1 make"
·
Apr 23, 2023
master-53c8434
53c84343
·
Improve AVX2 for vec_dot_q4_3_q8_0 (#1138)
·
Apr 23, 2023
master-c9e2c26
c9e2c26f
·
A better `packNibbles` and `mul_sum_i8_pairs_float` implementation using AVX512 (#1119)
·
Apr 23, 2023
master-0e018fe
0e018fe0
·
ggml : fix Q4_3 cuBLAS
·
Apr 22, 2023
master-857308d
857308d1
·
ci : trigger CI for drafts, but not most PR actions (#1125)
·
Apr 22, 2023
master-c50b628
c50b6288
·
Fix CI: ARM NEON, quantization unit tests, editorconfig (#1122)
·
Apr 22, 2023
master-36b4f7e
36b4f7e0
·
llama : print timings on ctrl+c exit (#1021)
·
Apr 22, 2023
master-10f19c1
10f19c11
·
llama : have n_batch default to 512 (#1091)
·
Apr 22, 2023
master-7e312f1
7e312f16
·
cmake : fix build under Windows when enable BUILD_SHARED_LIBS (#1100)
·
Apr 22, 2023
master-872c365
872c365a
·
ggml : fix AVX build + update to new Q8_0 format
·
Apr 22, 2023
master-c5aa5e5
c5aa5e57
·
ggml : AVX2 optimization for vec_dot_q4_3_q8_0 and refactoring (#1099)
·
Apr 22, 2023
master-b6e7f9b
b6e7f9b0
·
llama : add api for getting/setting the complete state: rng, logits, embedding and kv_cache (#1105)
·
Apr 22, 2023
master-50cb666
50cb666b
·
Improve cuBLAS performance by using a memory pool (#1094)
·
Apr 21, 2023
1
…
24
25
26
27
28
29
30
31
32
…
38