prithivMLmods commited on
Commit
7e4084a
·
verified ·
1 Parent(s): f3e6e6c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +42 -3
README.md CHANGED
@@ -1,3 +1,42 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ language:
4
+ - en
5
+ base_model:
6
+ - POLARIS-Project/Polaris-1.7B-Preview
7
+ pipeline_tag: text-generation
8
+ library_name: transformers
9
+ tags:
10
+ - text-generation-inference
11
+ ---
12
+
13
+ # **Polaris-1.7B-Preview-f32-GGUF**
14
+
15
+ > Polaris is an open-source post-training method that uses reinforcement learning (RL) scaling to refine and enhance models with advanced reasoning abilities. Our research shows that even top-tier models like Qwen3-4B can achieve significant improvements on challenging reasoning tasks when optimized with Polaris. By leveraging open-source data and academic-level resources, Polaris pushes the capabilities of open-recipe reasoning models to unprecedented heights.
16
+
17
+ ## Model Files
18
+
19
+ | File Name | Quant Type | Size |
20
+ |-----------|------------|------|
21
+ | Polaris-1.7B-Preview.BF16.gguf | BF16 | 3.45 GB |
22
+ | Polaris-1.7B-Preview.F16.gguf | F16 | 3.45 GB |
23
+ | Polaris-1.7B-Preview.F32.gguf | F32 | 6.89 GB |
24
+ | Polaris-1.7B-Preview.Q2_K.gguf | Q2_K | 778 MB |
25
+ | Polaris-1.7B-Preview.Q3_K_L.gguf | Q3_K_L | 1 GB |
26
+ | Polaris-1.7B-Preview.Q3_K_M.gguf | Q3_K_M | 940 MB |
27
+ | Polaris-1.7B-Preview.Q3_K_S.gguf | Q3_K_S | 867 MB |
28
+ | Polaris-1.7B-Preview.Q4_K_M.gguf | Q4_K_M | 1.11 GB |
29
+ | Polaris-1.7B-Preview.Q4_K_S.gguf | Q4_K_S | 1.06 GB |
30
+ | Polaris-1.7B-Preview.Q5_K_M.gguf | Q5_K_M | 1.26 GB |
31
+ | Polaris-1.7B-Preview.Q5_K_S.gguf | Q5_K_S | 1.23 GB |
32
+ | Polaris-1.7B-Preview.Q6_K.gguf | Q6_K | 1.42 GB |
33
+ | Polaris-1.7B-Preview.Q8_0.gguf | Q8_0 | 1.83 GB |
34
+
35
+ ## Quants Usage
36
+
37
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
38
+
39
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
40
+ types (lower is better):
41
+
42
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)