Commit graph

13 commits

Author SHA1 Message Date
Georgi Gerganov
574406dc7e
ggml : add Q5_0 and Q5_1 quantization (#1187)
* ggml : add Q5_0 quantization (cuBLAS only)

* ggml : fix Q5_0 qh -> uint32_t

* ggml : fix q5_0 histogram stats

* ggml : q5_0 scalar dot product

* ggml : q5_0 ARM NEON dot

* ggml : q5_0 more efficient ARM NEON using uint64_t masks

* ggml : rename Q5_0 -> Q5_1

* ggml : adding Q5_0 mode

* quantize : add Q5_0 and Q5_1 to map

* ggml : AVX2 optimizations for Q5_0, Q5_1 (#1195)

---------

Co-authored-by: Stephan Walter <stephan@walter.name>
2023-04-26 23:14:13 +03:00
Georgi Gerganov
884e7d7a2b
ggml : use 8-bit precision for Q4_1 intermediate results (#1047)
* ggml : use 8-bit precision for Q4_1 intermediate results (ARM)

* ggml : optimize ggml_vec_dot_q4_1_q8_0() via vmalq_n_f32

56 ms/token with Q4_1 !

* ggml : AVX2 implementation of ggml_vec_dot_q4_1_q8_0 (#1051)

* gitignore : ignore ppl-*.txt files

---------

Co-authored-by: slaren <2141330+slaren@users.noreply.github.com>
2023-04-19 20:10:08 +03:00
Georgi Gerganov
4caebf6d40
gitignore : vdot 2023-04-18 23:00:08 +03:00
Georgi Gerganov
c85980acd0
gitignore : benchmark 2023-04-13 18:01:33 +03:00
unbounded
62cfc54f77
Add quantize-stats command for testing quantization (#728)
Command that calculates some statistics over the errors introduced by
quantization, like mean square error, max error and some percentile errors for layer
weights. Should be useful for testing quantization improvements.

Exposes some internal state from ggml and llama for testing
2023-04-08 00:09:18 +02:00
iacore
ed1c214e66
zig : add build.zig (#773)
Co-authored-by: Locria Cyber <74560659+locriacyber@users.noreply.github.com>
2023-04-05 18:06:02 +03:00
Justine Tunney
78ca9838ee Make loading weights 10-100x faster
This is a breaking change that's going to give you three benefits:

1. Your inference commands should load 100x faster
2. You may be able to safely load models 2x larger
3. You can run many concurrent inference processes

This was accomplished by changing the file format so we can mmap()
weights directly into memory without having to read() or copy them
thereby ensuring the kernel can make its file cache pages directly
accessible to our inference processes; and secondly, that the file
cache pages are much less likely to get evicted (which would force
loads to hit disk) because they're no longer competing with memory
pages that were needlessly created by gigabytes of standard i/o.

The new file format supports single-file models like LLaMA 7b, and
it also supports multi-file models like LLaMA 13B. Our Python tool
now merges the foo.1, foo.2, etc. files back into a single file so
that the C++ code which maps it doesn't need to reshape data every
time. That's made llama.cpp so much simpler. Much of its load code
has now been deleted.

Furthermore, this change ensures that tensors are aligned properly
on a 32-byte boundary. That opens the door to seeing if we can get
additional performance gains on some microprocessors, by using ops
that require memory alignment.

Lastly note that both POSIX and the Windows platform are supported

Fixes #91
2023-03-30 12:28:25 -07:00
Jed Fox
20e1e84884
deploy : add a Package.swift for SwiftPM support (#393)
* Add a Package.swift for SwiftPM support

* Swap from exclusions to allowlist
2023-03-28 19:39:01 +03:00
Georgi Gerganov
e0670260fb
gitignore : add "embedding" 2023-03-28 18:34:35 +03:00
Georgi Gerganov
a316a425d0
Overhaul the examples structure
- main -> examples
- utils -> examples (renamed to "common")
- quantize -> examples
- separate tools for "perplexity" and "embedding"

Hope I didn't break something !
2023-03-25 20:26:40 +02:00
Niklas Korz
a292747893
Nix flake (#40)
* Nix flake

* Nix: only add Accelerate framework on macOS

* Nix: development shel, direnv and compatibility

* Nix: use python packages supplied by withPackages

* Nix: remove channel compatibility

* Nix: fix ARM neon dotproduct on macOS

---------

Co-authored-by: Pavol Rusnak <pavol@rusnak.io>
2023-03-17 23:03:48 +01:00
Georgi Gerganov
f60fa9e50a
.gitignore models/ 2023-03-11 12:27:02 +02:00
Georgi Gerganov
26c0846629
Initial release 2023-03-10 20:56:40 +02:00