Update README.md
Browse files
README.md
CHANGED
|
@@ -32,7 +32,11 @@ Let us know what you think of the model! The 8B and 12B versions of RPMax had gr
|
|
| 32 |
|
| 33 |
The model is available in quantized formats:
|
| 34 |
|
|
|
|
|
|
|
| 35 |
* **FP16**: https://huggingface.co/ArliAI/Llama-3.1-70B-ArliAI-RPMax-v1.1
|
|
|
|
|
|
|
| 36 |
* **GGUF**: https://huggingface.co/ArliAI/Llama-3.1-70B-ArliAI-RPMax-v1.1-GGUF
|
| 37 |
|
| 38 |
## Suggested Prompt Format
|
|
|
|
| 32 |
|
| 33 |
The model is available in quantized formats:
|
| 34 |
|
| 35 |
+
We recommend using full weights or GPTQ as GGUF seems to generate gibberish at low quants.
|
| 36 |
+
|
| 37 |
* **FP16**: https://huggingface.co/ArliAI/Llama-3.1-70B-ArliAI-RPMax-v1.1
|
| 38 |
+
* **GPTQ_Q4**: https://huggingface.co/ArliAI/Llama-3.1-70B-ArliAI-RPMax-v1.1-GPTQ_Q4
|
| 39 |
+
* **GPTQ_Q8**: https://huggingface.co/ArliAI/Llama-3.1-70B-ArliAI-RPMax-v1.1-GPTQ_Q8
|
| 40 |
* **GGUF**: https://huggingface.co/ArliAI/Llama-3.1-70B-ArliAI-RPMax-v1.1-GGUF
|
| 41 |
|
| 42 |
## Suggested Prompt Format
|