open-llama-3b 600bt preview ggml files
Browse files- README.md +19 -0
- SHA256SUMS +6 -0
- open-llama-3b-f16.bin +3 -0
- open-llama-3b-q4-0.bin +3 -0
- open-llama-3b-q4-1.bin +3 -0
- open-llama-3b-q5-0.bin +3 -0
- open-llama-3b-q5-1.bin +3 -0
- open-llama-3b-q8-0.bin +3 -0
README.md
CHANGED
|
@@ -1,3 +1,22 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
# ggml versions of OpenLLaMa 3B
|
| 6 |
+
|
| 7 |
+
- Version: 600b token preview
|
| 8 |
+
- Project: [OpenLLaMA: An Open Reproduction of LLaMA](https://github.com/openlm-research/open_llama)
|
| 9 |
+
- Model: [openlm-research/open_llama_3b_600bt_preview](https://huggingface.co/openlm-research/open_llama_3b_600bt_preview)
|
| 10 |
+
- llama.cpp: [#1588](https://github.com/ggerganov/llama.cpp/pull/1588)
|
| 11 |
+
|
| 12 |
+
## Use with llama.cpp
|
| 13 |
+
|
| 14 |
+
Since it is experimental and not merged yet, you can use the PR:
|
| 15 |
+
|
| 16 |
+
```sh
|
| 17 |
+
git clone https://github.com/ggerganov/llama.cpp
|
| 18 |
+
cd llama.cpp
|
| 19 |
+
curl -L https://github.com/ggerganov/llama.cpp/pull/1588.diff | git apply -
|
| 20 |
+
make main
|
| 21 |
+
./main -m open-llama-3b-q4-0.bin ...
|
| 22 |
+
```
|
SHA256SUMS
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
4461ccd289eed0190045fa79447262401fe432b63e6d9a7919637c420814e90b open-llama-3b-f16.bin
|
| 2 |
+
d4d4f2425f355dd57cae7c6766bbd99cf482c8b374cbf775c230f1a8c038c617 open-llama-3b-q8-0.bin
|
| 3 |
+
0103204cb367a4ae78a6dcc107ee95a0f0f216e6d276082a534e0dc337dd7452 open-llama-3b-q5-1.bin
|
| 4 |
+
7ed15048e392ce43abae56668f8df6cb0f7f1d48e4c8e924a9fc58a82510e6ac open-llama-3b-q5-0.bin
|
| 5 |
+
6e3b1e60f3135395bd32d8bb10388051c24b79bc5c0b5bc5e9cab11ebea253c3 open-llama-3b-q4-1.bin
|
| 6 |
+
878a64232542f174ecd41ca76f18b959cdf41944fb878b5cf6cb89ab264bd59b open-llama-3b-q4-0.bin
|
open-llama-3b-f16.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4461ccd289eed0190045fa79447262401fe432b63e6d9a7919637c420814e90b
|
| 3 |
+
size 6853758208
|
open-llama-3b-q4-0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:878a64232542f174ecd41ca76f18b959cdf41944fb878b5cf6cb89ab264bd59b
|
| 3 |
+
size 1928446208
|
open-llama-3b-q4-1.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6e3b1e60f3135395bd32d8bb10388051c24b79bc5c0b5bc5e9cab11ebea253c3
|
| 3 |
+
size 2142590208
|
open-llama-3b-q5-0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ed15048e392ce43abae56668f8df6cb0f7f1d48e4c8e924a9fc58a82510e6ac
|
| 3 |
+
size 2356734208
|
open-llama-3b-q5-1.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0103204cb367a4ae78a6dcc107ee95a0f0f216e6d276082a534e0dc337dd7452
|
| 3 |
+
size 2570878208
|
open-llama-3b-q8-0.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4d4f2425f355dd57cae7c6766bbd99cf482c8b374cbf775c230f1a8c038c617
|
| 3 |
+
size 3641598208
|