Hugging Face's logo

mradermacher
/
SRA-LLM-i1-GGUF

Transformers
GGUF
English
text-generation-inference
unsloth
gpt_oss
sarcasm
smart
reasoning
Model card Files
xet
Community
mradermacher commited on
Commit
2ae3e48
·
verified ·
1 Parent(s): fe11023

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +83 -0
README.md CHANGED
@@ -1,3 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  <!-- ### quantize_version: 2 -->
2
  <!-- ### output_tensor_quantised: 1 -->
3
  <!-- ### convert_type: hf -->
@@ -7,3 +27,66 @@
7
  <!-- ### quants_skip: -->
8
  <!-- ### skip_mmproj: -->
9
  weighted/imatrix quants of https://huggingface.co/Daemontatox/SRA-LLM
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Daemontatox/SRA-LLM
3
+ language:
4
+ - en
5
+ library_name: transformers
6
+ license: apache-2.0
7
+ mradermacher:
8
+ readme_rev: 1
9
+ quantized_by: mradermacher
10
+ tags:
11
+ - text-generation-inference
12
+ - transformers
13
+ - unsloth
14
+ - gpt_oss
15
+ - sarcasm
16
+ - smart
17
+ - reasoning
18
+ ---
19
+ ## About
20
+
21
  <!-- ### quantize_version: 2 -->
22
  <!-- ### output_tensor_quantised: 1 -->
23
  <!-- ### convert_type: hf -->
 
27
  <!-- ### quants_skip: -->
28
  <!-- ### skip_mmproj: -->
29
  weighted/imatrix quants of https://huggingface.co/Daemontatox/SRA-LLM
30
+
31
+ <!-- provided-files -->
32
+
33
+ ***For a convenient overview and download list, visit our [model page for this model](https://hf.tst.eu/model#SRA-LLM-i1-GGUF).***
34
+
35
+ static quants are available at https://huggingface.co/mradermacher/SRA-LLM-GGUF
36
+ ## Usage
37
+
38
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
39
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
40
+ more details, including on how to concatenate multi-part files.
41
+
42
+ ## Provided Quants
43
+
44
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
45
+
46
+ | Link | Type | Size/GB | Notes |
47
+ |:-----|:-----|--------:|:------|
48
+ | [GGUF](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.imatrix.gguf) | imatrix | 0.3 | imatrix file (for creating your own qwuants) |
49
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ1_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ1_M.gguf.part2of2) | i1-IQ1_M | 66.2 | mostly desperate |
50
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ1_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ1_S.gguf.part2of2) | i1-IQ1_S | 66.2 | for the desperate |
51
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ2_XXS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ2_XXS.gguf.part2of2) | i1-IQ2_XXS | 66.2 | |
52
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ2_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ2_XS.gguf.part2of2) | i1-IQ2_XS | 66.2 | |
53
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q3_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q3_K_S.gguf.part2of2) | i1-Q3_K_S | 66.2 | IQ3_XS probably better |
54
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ2_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ2_M.gguf.part2of2) | i1-IQ2_M | 66.3 | |
55
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ2_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ2_S.gguf.part2of2) | i1-IQ2_S | 66.3 | |
56
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ3_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ3_S.gguf.part2of2) | i1-IQ3_S | 66.3 | beats Q3_K* |
57
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ3_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ3_XS.gguf.part2of2) | i1-IQ3_XS | 66.3 | |
58
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ3_XXS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ3_XXS.gguf.part2of2) | i1-IQ3_XXS | 66.3 | lower quality |
59
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q2_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q2_K.gguf.part2of2) | i1-Q2_K | 66.3 | IQ3_XXS probably better |
60
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ4_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ4_XS.gguf.part2of2) | i1-IQ4_XS | 66.3 | |
61
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q4_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q4_0.gguf.part2of2) | i1-Q4_0 | 66.6 | fast, low quality |
62
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q2_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q2_K_S.gguf.part2of2) | i1-Q2_K_S | 66.7 | very low quality |
63
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ3_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-IQ3_M.gguf.part2of2) | i1-IQ3_M | 66.8 | |
64
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q3_K_M.gguf.part2of2) | i1-Q3_K_M | 71.2 | IQ3_S probably better |
65
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q3_K_L.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q3_K_L.gguf.part2of2) | i1-Q3_K_L | 73.5 | IQ3_M probably better |
66
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q4_1.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q4_1.gguf.part2of2) | i1-Q4_1 | 73.6 | |
67
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q4_K_S.gguf.part2of2) | i1-Q4_K_S | 81.0 | optimal size/speed/quality |
68
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q4_K_M.gguf.part2of2) | i1-Q4_K_M | 88.0 | fast, recommended |
69
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q5_K_S.gguf.part2of2) | i1-Q5_K_S | 88.1 | |
70
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q5_K_M.gguf.part2of2) | i1-Q5_K_M | 94.0 | |
71
+ | [PART 1](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q6_K.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q6_K.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/SRA-LLM-i1-GGUF/resolve/main/SRA-LLM.i1-Q6_K.gguf.part3of3) | i1-Q6_K | 124.3 | practically like static Q6_K |
72
+
73
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
74
+ types (lower is better):
75
+
76
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
77
+
78
+ And here are Artefact2's thoughts on the matter:
79
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
80
+
81
+ ## FAQ / Model Request
82
+
83
+ See https://huggingface.co/mradermacher/model_requests for some answers to
84
+ questions you might have and/or if you want some other model quantized.
85
+
86
+ ## Thanks
87
+
88
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
89
+ me use its servers and providing upgrades to my workstation to enable
90
+ this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
91
+
92
+ <!-- end -->