aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorGeorgi Gerganov <ggerganov@gmail.com>2023-03-11 11:34:11 +0200
committerGitHub <noreply@github.com>2023-03-11 11:34:11 +0200
commitea977e85ecda7b983f0e7b1db20b509998ddc889 (patch)
tree4ab35368de5aadb63d35b070e426e031d739df77
parent007a8f6f459c6eb56678fdee4c09219ddb85b640 (diff)
Update README.md
-rw-r--r--README.md6
1 files changed, 6 insertions, 0 deletions
diff --git a/README.md b/README.md
index 02b0c09..e645172 100644
--- a/README.md
+++ b/README.md
@@ -2,9 +2,15 @@
Inference of [Facebook's LLaMA](https://github.com/facebookresearch/llama) model in pure C/C++
+**!!! IMPORTANT!!!**
+
+Commit [007a8f6f459c6eb56678fdee4c09219ddb85b640](https://github.com/ggerganov/llama.cpp/commit/007a8f6f459c6eb56678fdee4c09219ddb85b640) added support for all LLaMA models, but introduced breaking changes. If you generated any models before that commit, you must regenerate them after updating to latest master.
+
+
**TEMPORARY NOTICE:**
Currently the quantized models run **only** on Apple Silicon. On other architectures, you can [use the F16 models](https://github.com/ggerganov/llama.cpp/issues/2#issuecomment-1464615286), but they will be much slower. Support will be [added later](https://github.com/ggerganov/ggml/pull/27)
+
## Description
The main goal is to run the model using 4-bit quantization on a MacBook.