add README
Browse files
README.md
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: gguf
|
| 3 |
+
tags:
|
| 4 |
+
- llama.cpp
|
| 5 |
+
- test-model
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
# stories15M
|
| 9 |
+
|
| 10 |
+
Tiny model used by the [llama.cpp](https://github.com/ggml-org/llama.cpp) test suite and CI. It is **not** meant for real-world use.
|
| 11 |
+
|
| 12 |
+
Originally hosted at `ggml-org/models/tinyllamas/`.
|
| 13 |
+
|
| 14 |
+
- `stories15M*.gguf`: [karpathy/tinyllamas](https://huggingface.co/karpathy/tinyllamas) stories15M (F32, Q4_0, Q8_0)
|
| 15 |
+
- `stories15M-be*.gguf`: big-endian variants (s390x CI)
|
| 16 |
+
- `split/`: the same model split into 3 shards (F32 and Q8_0), for split-loading tests
|