Update README.md

README.md

@@ -1,3 +1,43 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+language:
+- en
+base_model:
+- ValiantLabs/Qwen3-1.7B-ShiningValiant3
+pipeline_tag: text-generation
+library_name: transformers
+tags:
+- text-generation-inference
+---
+
+# **Qwen3-1.7B-ShiningValiant3-f32-GGUF**
+
+> Shining Valiant 3—available in Qwen3-1.7B, Qwen3-4B, and Qwen3-8B variants—is a specialized model focused on science, AI design, and general reasoning, built on the Qwen 3 architecture. It is fine-tuned on our latest high-difficulty science reasoning dataset generated with Deepseek R1 0528, leveraging AI to build better AI. This enhancement enables Shining Valiant 3 to assist in exploring cutting-edge innovations and improvements in AI. With improved general and creative reasoning, it excels in problem-solving and everyday conversational tasks. Its compact size ensures compatibility with local desktops, mobile devices, and ultra-fast server-side inference.
+
+
+## Model Files
+
+| File Name | Size | Quant Type |
+|-----------|------|------------|
+| Qwen3-1.7B-ShiningValiant3.F32.gguf | 6.89 GB | F32 |
+| Qwen3-1.7B-ShiningValiant3.BF16.gguf | 3.45 GB | BF16 |
+| Qwen3-1.7B-ShiningValiant3.F16.gguf | 3.45 GB | F16 |
+| Qwen3-1.7B-ShiningValiant3.Q8_0.gguf | 1.83 GB | Q8_0 |
+| Qwen3-1.7B-ShiningValiant3.Q6_K.gguf | 1.42 GB | Q6_K |
+| Qwen3-1.7B-ShiningValiant3.Q5_K_M.gguf | 1.26 GB | Q5_K_M |
+| Qwen3-1.7B-ShiningValiant3.Q5_K_S.gguf | 1.23 GB | Q5_K_S |
+| Qwen3-1.7B-ShiningValiant3.Q4_K_M.gguf | 1.11 GB | Q4_K_M |
+| Qwen3-1.7B-ShiningValiant3.Q4_K_S.gguf | 1.06 GB | Q4_K_S |
+| Qwen3-1.7B-ShiningValiant3.Q3_K_L.gguf | 1 GB | Q3_K_L |
+| Qwen3-1.7B-ShiningValiant3.Q3_K_M.gguf | 940 MB | Q3_K_M |
+| Qwen3-1.7B-ShiningValiant3.Q3_K_S.gguf | 867 MB | Q3_K_S |
+| Qwen3-1.7B-ShiningValiant3.Q2_K.gguf | 778 MB | Q2_K |
+
+## Quants Usage 
+
+(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
+
+Here is a handy graph by ikawrakow comparing some lower-quality quant
+types (lower is better):
+
+![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)