Hugging Face's logo

mradermacher
/
gpt-oss-20b-DAG-Reasoning-i1-GGUF

Transformers
GGUF
English
dag-reasoning
gpt
gpt-oss
gpt-oss-20b
openai
20b
reasoning
directed-acyclic-graph
graph
logic
analysis
programming
knowledge
root-cause-analysis
economics
business
business-management
finance
law
supply-chain
logistics
software-engineering
cybersecurity
architecture
energy
politics
problem-solving
creative
analytical
expert
rationality
conversational
chat
instruct
imatrix
Model card Files
xet
Community
mradermacher commited on
Commit
7f7a603
·
verified ·
1 Parent(s): 44fcb1f

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +112 -0
README.md CHANGED
@@ -1,3 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  <!-- ### quantize_version: 2 -->
2
  <!-- ### output_tensor_quantised: 1 -->
3
  <!-- ### convert_type: hf -->
@@ -7,3 +56,66 @@
7
  <!-- ### quants_skip: -->
8
  <!-- ### skip_mmproj: -->
9
  weighted/imatrix quants of https://huggingface.co/sequelbox/gpt-oss-20b-DAG-Reasoning
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: sequelbox/gpt-oss-20b-DAG-Reasoning
3
+ datasets:
4
+ - sequelbox/DAG-Reasoning-DeepSeek-R1-0528
5
+ language:
6
+ - en
7
+ library_name: transformers
8
+ license: apache-2.0
9
+ mradermacher:
10
+ readme_rev: 1
11
+ quantized_by: mradermacher
12
+ tags:
13
+ - dag-reasoning
14
+ - gpt
15
+ - gpt-oss
16
+ - gpt-oss-20b
17
+ - openai
18
+ - 20b
19
+ - reasoning
20
+ - directed-acyclic-graph
21
+ - graph
22
+ - logic
23
+ - analysis
24
+ - programming
25
+ - knowledge
26
+ - root-cause-analysis
27
+ - economics
28
+ - business
29
+ - business-management
30
+ - finance
31
+ - law
32
+ - supply-chain
33
+ - logistics
34
+ - software-engineering
35
+ - cybersecurity
36
+ - architecture
37
+ - energy
38
+ - politics
39
+ - problem-solving
40
+ - creative
41
+ - analytical
42
+ - expert
43
+ - rationality
44
+ - conversational
45
+ - chat
46
+ - instruct
47
+ ---
48
+ ## About
49
+
50
  <!-- ### quantize_version: 2 -->
51
  <!-- ### output_tensor_quantised: 1 -->
52
  <!-- ### convert_type: hf -->
 
56
  <!-- ### quants_skip: -->
57
  <!-- ### skip_mmproj: -->
58
  weighted/imatrix quants of https://huggingface.co/sequelbox/gpt-oss-20b-DAG-Reasoning
59
+
60
+ <!-- provided-files -->
61
+
62
+ ***For a convenient overview and download list, visit our [model page for this model](https://hf.tst.eu/model#gpt-oss-20b-DAG-Reasoning-i1-GGUF).***
63
+
64
+ static quants are available at https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-GGUF
65
+ ## Usage
66
+
67
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
68
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
69
+ more details, including on how to concatenate multi-part files.
70
+
71
+ ## Provided Quants
72
+
73
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
74
+
75
+ | Link | Type | Size/GB | Notes |
76
+ |:-----|:-----|--------:|:------|
77
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
78
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ1_M.gguf) | i1-IQ1_M | 12.1 | mostly desperate |
79
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ1_S.gguf) | i1-IQ1_S | 12.1 | for the desperate |
80
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 12.1 | |
81
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ2_XS.gguf) | i1-IQ2_XS | 12.1 | |
82
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q3_K_S.gguf) | i1-Q3_K_S | 12.2 | IQ3_XS probably better |
83
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ2_M.gguf) | i1-IQ2_M | 12.2 | |
84
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ2_S.gguf) | i1-IQ2_S | 12.2 | |
85
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ3_S.gguf) | i1-IQ3_S | 12.2 | beats Q3_K* |
86
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ3_XS.gguf) | i1-IQ3_XS | 12.2 | |
87
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 12.2 | lower quality |
88
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q2_K.gguf) | i1-Q2_K | 12.2 | IQ3_XXS probably better |
89
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ4_XS.gguf) | i1-IQ4_XS | 12.2 | |
90
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q2_K_S.gguf) | i1-Q2_K_S | 12.2 | very low quality |
91
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q4_0.gguf) | i1-Q4_0 | 12.2 | fast, low quality |
92
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-IQ3_M.gguf) | i1-IQ3_M | 12.3 | |
93
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q3_K_M.gguf) | i1-Q3_K_M | 13.0 | IQ3_S probably better |
94
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q3_K_L.gguf) | i1-Q3_K_L | 13.4 | IQ3_M probably better |
95
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q4_1.gguf) | i1-Q4_1 | 13.5 | |
96
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q4_K_S.gguf) | i1-Q4_K_S | 14.8 | optimal size/speed/quality |
97
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q4_K_M.gguf) | i1-Q4_K_M | 15.9 | fast, recommended |
98
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q5_K_S.gguf) | i1-Q5_K_S | 16.0 | |
99
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q5_K_M.gguf) | i1-Q5_K_M | 17.0 | |
100
+ | [GGUF](https://huggingface.co/mradermacher/gpt-oss-20b-DAG-Reasoning-i1-GGUF/resolve/main/gpt-oss-20b-DAG-Reasoning.i1-Q6_K.gguf) | i1-Q6_K | 22.3 | practically like static Q6_K |
101
+
102
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
103
+ types (lower is better):
104
+
105
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
106
+
107
+ And here are Artefact2's thoughts on the matter:
108
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
109
+
110
+ ## FAQ / Model Request
111
+
112
+ See https://huggingface.co/mradermacher/model_requests for some answers to
113
+ questions you might have and/or if you want some other model quantized.
114
+
115
+ ## Thanks
116
+
117
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
118
+ me use its servers and providing upgrades to my workstation to enable
119
+ 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.
120
+
121
+ <!-- end -->