Hugging Face's logo

mradermacher
/
Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF

Transformers
GGUF
English
Inference Endpoints
imatrix
conversational
Model card Files Community
mradermacher commited on
Commit
d12fbdd
·
verified ·
1 Parent(s): 8f106ef

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +81 -0
README.md CHANGED
@@ -1,6 +1,87 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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/meditsolutions/Llama-3.2-SUN-HDIC-1B-Instruct
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: meditsolutions/Llama-3.2-SUN-HDIC-1B-Instruct
3
+ datasets:
4
+ - argilla/OpenHermesPreferences
5
+ - argilla/magpie-ultra-v0.1
6
+ - argilla/Capybara-Preferences-Filtered
7
+ - mlabonne/open-perfectblend
8
+ - HuggingFaceTB/everyday-conversations-llama3.1-2k
9
+ - WizardLMTeam/WizardLM_evol_instruct_V2_196k
10
+ - ProlificAI/social-reasoning-rlhf
11
+ - allenai/tulu-3-sft-mixture
12
+ - allenai/llama-3.1-tulu-3-8b-preference-mixture
13
+ language:
14
+ - en
15
+ library_name: transformers
16
+ license: llama3.2
17
+ quantized_by: mradermacher
18
+ ---
19
+ ## About
20
+
21
  <!-- ### quantize_version: 2 -->
22
  <!-- ### output_tensor_quantised: 1 -->
23
  <!-- ### convert_type: hf -->
24
  <!-- ### vocab_type: -->
25
  <!-- ### tags: nicoboss -->
26
  weighted/imatrix quants of https://huggingface.co/meditsolutions/Llama-3.2-SUN-HDIC-1B-Instruct
27
+
28
+ <!-- provided-files -->
29
+ static quants are available at https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-GGUF
30
+ ## Usage
31
+
32
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
33
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
34
+ more details, including on how to concatenate multi-part files.
35
+
36
+ ## Provided Quants
37
+
38
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
39
+
40
+ | Link | Type | Size/GB | Notes |
41
+ |:-----|:-----|--------:|:------|
42
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ1_S.gguf) | i1-IQ1_S | 0.6 | for the desperate |
43
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ1_M.gguf) | i1-IQ1_M | 0.6 | mostly desperate |
44
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 0.6 | |
45
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ2_XS.gguf) | i1-IQ2_XS | 0.7 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ2_S.gguf) | i1-IQ2_S | 0.7 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ2_M.gguf) | i1-IQ2_M | 0.7 | |
48
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q2_K_S.gguf) | i1-Q2_K_S | 0.7 | very low quality |
49
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q2_K.gguf) | i1-Q2_K | 0.8 | IQ3_XXS probably better |
50
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 0.8 | lower quality |
51
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ3_XS.gguf) | i1-IQ3_XS | 0.8 | |
52
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q3_K_S.gguf) | i1-Q3_K_S | 0.9 | IQ3_XS probably better |
53
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ3_S.gguf) | i1-IQ3_S | 0.9 | beats Q3_K* |
54
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ3_M.gguf) | i1-IQ3_M | 0.9 | |
55
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q3_K_M.gguf) | i1-Q3_K_M | 0.9 | IQ3_S probably better |
56
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q3_K_L.gguf) | i1-Q3_K_L | 0.9 | IQ3_M probably better |
57
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-IQ4_XS.gguf) | i1-IQ4_XS | 1.0 | |
58
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 1.0 | fast on arm, low quality |
59
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 1.0 | fast on arm+i8mm, low quality |
60
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 1.0 | fast on arm+sve, low quality |
61
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q4_0.gguf) | i1-Q4_0 | 1.0 | fast, low quality |
62
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q4_K_S.gguf) | i1-Q4_K_S | 1.0 | optimal size/speed/quality |
63
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q4_K_M.gguf) | i1-Q4_K_M | 1.1 | fast, recommended |
64
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q5_K_S.gguf) | i1-Q5_K_S | 1.2 | |
65
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q5_K_M.gguf) | i1-Q5_K_M | 1.2 | |
66
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.2-SUN-HDIC-1B-Instruct-i1-GGUF/resolve/main/Llama-3.2-SUN-HDIC-1B-Instruct.i1-Q6_K.gguf) | i1-Q6_K | 1.3 | practically like static Q6_K |
67
+
68
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
69
+ types (lower is better):
70
+
71
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
72
+
73
+ And here are Artefact2's thoughts on the matter:
74
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
75
+
76
+ ## FAQ / Model Request
77
+
78
+ See https://huggingface.co/mradermacher/model_requests for some answers to
79
+ questions you might have and/or if you want some other model quantized.
80
+
81
+ ## Thanks
82
+
83
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
84
+ me use its servers and providing upgrades to my workstation to enable
85
+ 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.
86
+
87
+ <!-- end -->