mradermacher
commited on
Commit
•
78d82a8
1
Parent(s):
5c7cb37
auto-patch README.md
Browse files
README.md
CHANGED
@@ -35,7 +35,7 @@ more details, including on how to concatenate multi-part files.
|
|
35 |
| Link | Type | Size/GB | Notes |
|
36 |
|:-----|:-----|--------:|:------|
|
37 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-IQ1_S.gguf) | i1-IQ1_S | 9.7 | for the desperate |
|
38 |
-
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-IQ1_M.gguf) | i1-IQ1_M | 10.6 |
|
39 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 12.0 | |
|
40 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-IQ2_XS.gguf) | i1-IQ2_XS | 13.2 | |
|
41 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-IQ2_S.gguf) | i1-IQ2_S | 13.9 | |
|
@@ -56,7 +56,6 @@ more details, including on how to concatenate multi-part files.
|
|
56 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-Q5_K_M.gguf) | i1-Q5_K_M | 30.7 | |
|
57 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-Q6_K.gguf) | i1-Q6_K | 35.5 | practically like static Q6_K |
|
58 |
|
59 |
-
|
60 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
61 |
types (lower is better):
|
62 |
|
|
|
35 |
| Link | Type | Size/GB | Notes |
|
36 |
|:-----|:-----|--------:|:------|
|
37 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-IQ1_S.gguf) | i1-IQ1_S | 9.7 | for the desperate |
|
38 |
+
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-IQ1_M.gguf) | i1-IQ1_M | 10.6 | mostly desperate |
|
39 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 12.0 | |
|
40 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-IQ2_XS.gguf) | i1-IQ2_XS | 13.2 | |
|
41 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-IQ2_S.gguf) | i1-IQ2_S | 13.9 | |
|
|
|
56 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-Q5_K_M.gguf) | i1-Q5_K_M | 30.7 | |
|
57 |
| [GGUF](https://huggingface.co/mradermacher/llama3-42b-v0-i1-GGUF/resolve/main/llama3-42b-v0.i1-Q6_K.gguf) | i1-Q6_K | 35.5 | practically like static Q6_K |
|
58 |
|
|
|
59 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
60 |
types (lower is better):
|
61 |
|