Commit Graph

15 Commits

Author SHA1 Message Date
Justine Tunney
ede59bb742 Add BF16 support and fix warnings
This change updates the data type definitions to be the same as the
latest source code. Support for the bfloat16 data type is available
however it can't interpret the IQ quantization formats yet. Cleanup
of compiler warnings and other nits have been fixed, but behavioral
changes have been avoided, and no new features are as of yet added.
2024-05-25 22:58:50 -07:00
Jack Mousseau
7d25893516 Match key-value pair and tensor counts with header integer width 2024-02-18 07:26:35 -08:00
antirez
eec3dc9f54 F16 output for dequantization. 2024-01-09 18:46:26 +01:00
antirez
26e3a59233 Rename gguf_init/end to more obvious names. 2024-01-09 16:35:40 +01:00
antirez
6eb4aeb2fb gguf_create(): take flags to be able to overwrite files. Fixes #7. 2024-01-09 16:32:10 +01:00
Juarez Bochi
e5cdcec626 Fix some typos 2024-01-03 07:34:12 -05:00
antirez
e2062eea2c Q4_K dequantization. 2023-12-28 12:31:35 +01:00
antirez
c25ccfa02a Q8_0 dequantization. 2023-12-27 21:22:33 +01:00
antirez
bd4ecbda94 FP16 added. Split-mixtral improved. 2023-12-27 15:25:18 +01:00
antirez
a77a4d061c Mixtral experts extraction test. 2023-12-26 17:23:47 +01:00
antirez
3081d69b8e split-mixtral: copying of keys + APIs needed. 2023-12-26 09:14:50 +01:00
antirez
53fb176b3b Initial API to create new GGUF files.
Also added a few libraries that will be needed soon.
The CLI was renamed with the final name of gguf-tools.
2023-12-25 22:10:07 +01:00
antirez
3eb30c1872 API to remap/rewind + mapping in write mode. 2023-12-25 10:45:38 +01:00
antirez
b3092d3860 Compute tensor size in bytes. 2023-12-24 23:44:24 +01:00
antirez
d54409bc9c Some library layout. 2023-12-24 18:31:03 +01:00