Update README.md
Browse files
README.md
CHANGED
|
@@ -1,5 +1,3 @@
|
|
| 1 |
-
This is a 4-bit version of Sugoi 14B Ultra, quantized using GPTQmodel and the VNTL-v3.1-1k dataset. This quant should work better than GGUF for certain backends like vLLM and aphrodite-engine, which excel at asynchronous prompting.
|
| 2 |
-
|
| 3 |
---
|
| 4 |
license: apache-2.0
|
| 5 |
datasets:
|
|
@@ -14,6 +12,8 @@ pipeline_tag: text-generation
|
|
| 14 |
|
| 15 |
# Sugoi LLM 14B Ultra (HF version)
|
| 16 |
|
|
|
|
|
|
|
| 17 |
Unleashing the full potential of the previous sugoi 14B model, **Sugoi 14B Ultra** delivers near-double translation accuracy compared to its quantized predecessor—achieving a BLEU score of **21.38 vs 13.67**. Its prompt-following skills rival those of Qwen 2.5 Base, especially when handling the bracket-heavy text commonly found in RPG Maker projects.
|
| 18 |
|
| 19 |
---
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
datasets:
|
|
|
|
| 12 |
|
| 13 |
# Sugoi LLM 14B Ultra (HF version)
|
| 14 |
|
| 15 |
+
This is a 4-bit version of Sugoi 14B Ultra, quantized using GPTQmodel and the VNTL-v3.1-1k dataset. This quant should work better than GGUF for certain backends like vLLM and aphrodite-engine, which excel at asynchronous prompting.
|
| 16 |
+
|
| 17 |
Unleashing the full potential of the previous sugoi 14B model, **Sugoi 14B Ultra** delivers near-double translation accuracy compared to its quantized predecessor—achieving a BLEU score of **21.38 vs 13.67**. Its prompt-following skills rival those of Qwen 2.5 Base, especially when handling the bracket-heavy text commonly found in RPG Maker projects.
|
| 18 |
|
| 19 |
---
|