Hugging Face's logo

mradermacher
/
EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF

Transformers
GGUF
PyTorch
facebook
meta
llama
llama-3
Inference Endpoints
imatrix
conversational
Model card Files Community
mradermacher commited on
Commit
a620bf0
•
1 Parent(s): b3030a5

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +83 -0
README.md CHANGED
@@ -1,6 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  <!-- ### quantize_version: 2 -->
2
  <!-- ### output_tensor_quantised: 1 -->
3
  <!-- ### convert_type: hf -->
4
  <!-- ### vocab_type: -->
5
  <!-- ### tags: nicoboss -->
6
  weighted/imatrix quants of https://huggingface.co/AXCXEPT/EZO-Llama-3.2-3B-Instruct-dpoE
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: AXCXEPT/EZO-Llama-3.2-3B-Instruct-dpoE
3
+ language:
4
+ - en
5
+ - jp
6
+ - de
7
+ - fr
8
+ - it
9
+ - pt
10
+ - hi
11
+ - es
12
+ - th
13
+ library_name: transformers
14
+ license: llama3.2
15
+ quantized_by: mradermacher
16
+ tags:
17
+ - facebook
18
+ - meta
19
+ - pytorch
20
+ - llama
21
+ - llama-3
22
+ ---
23
+ ## About
24
+
25
  <!-- ### quantize_version: 2 -->
26
  <!-- ### output_tensor_quantised: 1 -->
27
  <!-- ### convert_type: hf -->
28
  <!-- ### vocab_type: -->
29
  <!-- ### tags: nicoboss -->
30
  weighted/imatrix quants of https://huggingface.co/AXCXEPT/EZO-Llama-3.2-3B-Instruct-dpoE
31
+
32
+ <!-- provided-files -->
33
+ ## Usage
34
+
35
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
36
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
37
+ more details, including on how to concatenate multi-part files.
38
+
39
+ ## Provided Quants
40
+
41
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
42
+
43
+ | Link | Type | Size/GB | Notes |
44
+ |:-----|:-----|--------:|:------|
45
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ1_S.gguf) | i1-IQ1_S | 1.0 | for the desperate |
46
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ1_M.gguf) | i1-IQ1_M | 1.0 | mostly desperate |
47
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 1.1 | |
48
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ2_XS.gguf) | i1-IQ2_XS | 1.2 | |
49
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ2_S.gguf) | i1-IQ2_S | 1.3 | |
50
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ2_M.gguf) | i1-IQ2_M | 1.3 | |
51
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 1.4 | lower quality |
52
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q2_K.gguf) | i1-Q2_K | 1.5 | IQ3_XXS probably better |
53
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ3_XS.gguf) | i1-IQ3_XS | 1.6 | |
54
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ3_S.gguf) | i1-IQ3_S | 1.6 | beats Q3_K* |
55
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q3_K_S.gguf) | i1-Q3_K_S | 1.6 | IQ3_XS probably better |
56
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ3_M.gguf) | i1-IQ3_M | 1.7 | |
57
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q3_K_M.gguf) | i1-Q3_K_M | 1.8 | IQ3_S probably better |
58
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q3_K_L.gguf) | i1-Q3_K_L | 1.9 | IQ3_M probably better |
59
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-IQ4_XS.gguf) | i1-IQ4_XS | 1.9 | |
60
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 2.0 | fast on arm, low quality |
61
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 2.0 | fast on arm+i8mm, low quality |
62
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 2.0 | fast on arm+sve, low quality |
63
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q4_0.gguf) | i1-Q4_0 | 2.0 | fast, low quality |
64
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q4_K_S.gguf) | i1-Q4_K_S | 2.0 | optimal size/speed/quality |
65
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q4_K_M.gguf) | i1-Q4_K_M | 2.1 | fast, recommended |
66
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q5_K_S.gguf) | i1-Q5_K_S | 2.4 | |
67
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q5_K_M.gguf) | i1-Q5_K_M | 2.4 | |
68
+ | [GGUF](https://huggingface.co/mradermacher/EZO-Llama-3.2-3B-Instruct-dpoE-i1-GGUF/resolve/main/EZO-Llama-3.2-3B-Instruct-dpoE.i1-Q6_K.gguf) | i1-Q6_K | 2.7 | practically like static Q6_K |
69
+
70
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
71
+ types (lower is better):
72
+
73
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
74
+
75
+ And here are Artefact2's thoughts on the matter:
76
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
77
+
78
+ ## FAQ / Model Request
79
+
80
+ See https://huggingface.co/mradermacher/model_requests for some answers to
81
+ questions you might have and/or if you want some other model quantized.
82
+
83
+ ## Thanks
84
+
85
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
86
+ me use its servers and providing upgrades to my workstation to enable
87
+ 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.
88
+
89
+ <!-- end -->