Tags

Tags give the ability to mark specific points in history as being important

b2058

3c0d25c4 · make: add nvcc info print (#5310) · Feb 03, 2024
b2057

3cc5ed35 · make: fix nvcc optimization flags for host code (#5309) · Feb 03, 2024
b2055

e920ed39 · Vulkan Intel Fixes, Optimizations and Debugging Flags (#5301) · Feb 03, 2024
b2054

52bb63c7 · refactor : switch to emplace_back to avoid extra object (#5291) · Feb 03, 2024
b2053

1ec3332a · YaRN : store rope scaling type as int32_t in memory (#5285) · Feb 03, 2024
b2051

a305dba8 · Fix im2col with 32fp (#5286) · Feb 03, 2024
b2050

19122117 · perplexity : fix KL divergence calculations on Windows (#5273) · Feb 02, 2024
b2047

b05102fe · Tidy ggml-sycl (#5261) · Feb 02, 2024
b2045

e805f0fa · [SYCL] get MAX_MEM_ALLOC from device property (#5270) · Feb 02, 2024
b2043

e1e72109 · llama : fix memory leak in llama_batch_free (#5252) · Feb 02, 2024
b2042

128dcbd3 · add --no-mmap in llama-bench (#5257) · Feb 01, 2024
b2041

4d0924a8 · Vulkan Phi Fix for AMD Proprietary Drivers (#5260) · Feb 01, 2024
b2040

8ca511ca · cuda : fix LLAMA_CUDA_F16 (#5262) · Feb 01, 2024
b2039

d71ac909 · make : generate .a library for static linking (#5205) · Feb 01, 2024
b2038

ce320601 · llama : support InternLM2 (#5184) · Feb 01, 2024
b2037

1cfb5372 · Fix broken Vulkan Cmake (properly) (#5230) · Jan 31, 2024
b2036

d3bac7d5 · llama : reorder build_orion() at correct place (#5118) · Jan 31, 2024
b2035

5cb04dbc · llama : remove LLAMA_MAX_DEVICES and LLAMA_SUPPORTS_GPU_OFFLOAD (#5240) · Jan 31, 2024
b2034

efb7bdbb · metal : add im2col F32 dst support (#5132) · Jan 31, 2024
b2033

15606309 · llava : add MobileVLM support (#5132) · Jan 31, 2024

Prev
1
…
28
29
30
31
32
33
34
35
36
…
99
Next