Update README.md
Browse files
README.md
CHANGED
|
@@ -23,6 +23,7 @@ Models with higher precision 4-8bit after calibration may show better quality th
|
|
| 23 |
<b><u>DISCLAIMER: Be aware that quantised models show reduced response quality and possible hallucinations!</u></b><br>
|
| 24 |
|
| 25 |
### Available quantization formats:
|
|
|
|
| 26 |
* **IQ2_XXS:** Lower quality, uses SOTA techniques to be usable.
|
| 27 |
* **IQ3_XXS:** Lower quality, new method with decent performance, comparable to Q3 quants.
|
| 28 |
* **IQ4_XS:** Decent quality, smaller than Q4_K_S with similar performance, recommended.
|
|
|
|
| 23 |
<b><u>DISCLAIMER: Be aware that quantised models show reduced response quality and possible hallucinations!</u></b><br>
|
| 24 |
|
| 25 |
### Available quantization formats:
|
| 26 |
+
* **IQ1_M:** (1.75bit) Extremely low quality, not recommended.
|
| 27 |
* **IQ2_XXS:** Lower quality, uses SOTA techniques to be usable.
|
| 28 |
* **IQ3_XXS:** Lower quality, new method with decent performance, comparable to Q3 quants.
|
| 29 |
* **IQ4_XS:** Decent quality, smaller than Q4_K_S with similar performance, recommended.
|