Hugging Face's logo

mradermacher
/
Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF

Transformers
GGUF
English
logic
argumentation
critical-thinking
argument-mapping
trl
sft
Inference Endpoints
imatrix
conversational
Model card Files Community
mradermacher commited on
Commit
e350335
·
verified ·
1 Parent(s): eced361

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +80 -0
README.md CHANGED
@@ -1,6 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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/DebateLabKIT/Llama-3.3-Argunaut-1-70B-SFT
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: DebateLabKIT/Llama-3.3-Argunaut-1-70B-SFT
3
+ datasets:
4
+ - DebateLabKIT/deepa2-conversations
5
+ - DebateLabKIT/deep-argmap-conversations
6
+ - allenai/tulu-3-sft-mixture
7
+ language:
8
+ - en
9
+ library_name: transformers
10
+ license: llama3.3
11
+ quantized_by: mradermacher
12
+ tags:
13
+ - logic
14
+ - argumentation
15
+ - critical-thinking
16
+ - argument-mapping
17
+ - trl
18
+ - sft
19
+ ---
20
+ ## About
21
+
22
  <!-- ### quantize_version: 2 -->
23
  <!-- ### output_tensor_quantised: 1 -->
24
  <!-- ### convert_type: hf -->
25
  <!-- ### vocab_type: -->
26
  <!-- ### tags: nicoboss -->
27
  weighted/imatrix quants of https://huggingface.co/DebateLabKIT/Llama-3.3-Argunaut-1-70B-SFT
28
+
29
+ <!-- provided-files -->
30
+ static quants are available at https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-GGUF
31
+ ## Usage
32
+
33
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
34
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
35
+ more details, including on how to concatenate multi-part files.
36
+
37
+ ## Provided Quants
38
+
39
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
40
+
41
+ | Link | Type | Size/GB | Notes |
42
+ |:-----|:-----|--------:|:------|
43
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ1_S.gguf) | i1-IQ1_S | 15.4 | for the desperate |
44
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ1_M.gguf) | i1-IQ1_M | 16.9 | mostly desperate |
45
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 19.2 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ2_XS.gguf) | i1-IQ2_XS | 21.2 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ2_S.gguf) | i1-IQ2_S | 22.3 | |
48
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ2_M.gguf) | i1-IQ2_M | 24.2 | |
49
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q2_K_S.gguf) | i1-Q2_K_S | 24.6 | very low quality |
50
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q2_K.gguf) | i1-Q2_K | 26.5 | IQ3_XXS probably better |
51
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 27.6 | lower quality |
52
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ3_XS.gguf) | i1-IQ3_XS | 29.4 | |
53
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ3_S.gguf) | i1-IQ3_S | 31.0 | beats Q3_K* |
54
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q3_K_S.gguf) | i1-Q3_K_S | 31.0 | IQ3_XS probably better |
55
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ3_M.gguf) | i1-IQ3_M | 32.0 | |
56
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q3_K_M.gguf) | i1-Q3_K_M | 34.4 | IQ3_S probably better |
57
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q3_K_L.gguf) | i1-Q3_K_L | 37.2 | IQ3_M probably better |
58
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-IQ4_XS.gguf) | i1-IQ4_XS | 38.0 | |
59
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q4_0.gguf) | i1-Q4_0 | 40.2 | fast, low quality |
60
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q4_K_S.gguf) | i1-Q4_K_S | 40.4 | optimal size/speed/quality |
61
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q4_K_M.gguf) | i1-Q4_K_M | 42.6 | fast, recommended |
62
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q4_1.gguf) | i1-Q4_1 | 44.4 | |
63
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q5_K_S.gguf) | i1-Q5_K_S | 48.8 | |
64
+ | [GGUF](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q5_K_M.gguf) | i1-Q5_K_M | 50.0 | |
65
+ | [PART 1](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Llama-3.3-Argunaut-1-70B-SFT-i1-GGUF/resolve/main/Llama-3.3-Argunaut-1-70B-SFT.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 58.0 | practically like static Q6_K |
66
+
67
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
68
+ types (lower is better):
69
+
70
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
71
+
72
+ And here are Artefact2's thoughts on the matter:
73
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
74
+
75
+ ## FAQ / Model Request
76
+
77
+ See https://huggingface.co/mradermacher/model_requests for some answers to
78
+ questions you might have and/or if you want some other model quantized.
79
+
80
+ ## Thanks
81
+
82
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
83
+ me use its servers and providing upgrades to my workstation to enable
84
+ 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.
85
+
86
+ <!-- end -->