mradermacher commited on
Commit
a067fcb
·
verified ·
1 Parent(s): 851b393

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +96 -0
README.md CHANGED
@@ -1,6 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  <!-- ### quantize_version: 2 -->
2
  <!-- ### output_tensor_quantised: 1 -->
3
  <!-- ### convert_type: hf -->
4
  <!-- ### vocab_type: -->
5
  <!-- ### tags: -->
6
  static quants of https://huggingface.co/ValiantLabs/Qwen3-4B-Esper3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: ValiantLabs/Qwen3-4B-Esper3
3
+ datasets:
4
+ - sequelbox/Titanium2.1-DeepSeek-R1
5
+ - sequelbox/Tachibana2-DeepSeek-R1
6
+ - sequelbox/Raiden-DeepSeek-R1
7
+ language:
8
+ - en
9
+ library_name: transformers
10
+ license: apache-2.0
11
+ quantized_by: mradermacher
12
+ tags:
13
+ - esper
14
+ - esper-3
15
+ - valiant
16
+ - valiant-labs
17
+ - qwen
18
+ - qwen-3
19
+ - qwen-3-4b
20
+ - 4b
21
+ - reasoning
22
+ - code
23
+ - code-instruct
24
+ - python
25
+ - javascript
26
+ - dev-ops
27
+ - jenkins
28
+ - terraform
29
+ - scripting
30
+ - powershell
31
+ - azure
32
+ - aws
33
+ - gcp
34
+ - cloud
35
+ - problem-solving
36
+ - architect
37
+ - engineer
38
+ - developer
39
+ - creative
40
+ - analytical
41
+ - expert
42
+ - rationality
43
+ - conversational
44
+ - chat
45
+ - instruct
46
+ ---
47
+ ## About
48
+
49
  <!-- ### quantize_version: 2 -->
50
  <!-- ### output_tensor_quantised: 1 -->
51
  <!-- ### convert_type: hf -->
52
  <!-- ### vocab_type: -->
53
  <!-- ### tags: -->
54
  static quants of https://huggingface.co/ValiantLabs/Qwen3-4B-Esper3
55
+
56
+ <!-- provided-files -->
57
+ weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion.
58
+ ## Usage
59
+
60
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
61
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
62
+ more details, including on how to concatenate multi-part files.
63
+
64
+ ## Provided Quants
65
+
66
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
67
+
68
+ | Link | Type | Size/GB | Notes |
69
+ |:-----|:-----|--------:|:------|
70
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.Q2_K.gguf) | Q2_K | 1.8 | |
71
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.Q3_K_S.gguf) | Q3_K_S | 2.0 | |
72
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.Q3_K_M.gguf) | Q3_K_M | 2.2 | lower quality |
73
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.Q3_K_L.gguf) | Q3_K_L | 2.3 | |
74
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.IQ4_XS.gguf) | IQ4_XS | 2.4 | |
75
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.Q4_K_S.gguf) | Q4_K_S | 2.5 | fast, recommended |
76
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.Q4_K_M.gguf) | Q4_K_M | 2.6 | fast, recommended |
77
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.Q5_K_S.gguf) | Q5_K_S | 2.9 | |
78
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.Q5_K_M.gguf) | Q5_K_M | 3.0 | |
79
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.Q6_K.gguf) | Q6_K | 3.4 | very good quality |
80
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.Q8_0.gguf) | Q8_0 | 4.4 | fast, best quality |
81
+ | [GGUF](https://huggingface.co/mradermacher/Qwen3-4B-Esper3-GGUF/resolve/main/Qwen3-4B-Esper3.f16.gguf) | f16 | 8.2 | 16 bpw, overkill |
82
+
83
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
84
+ types (lower is better):
85
+
86
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
87
+
88
+ And here are Artefact2's thoughts on the matter:
89
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
90
+
91
+ ## FAQ / Model Request
92
+
93
+ See https://huggingface.co/mradermacher/model_requests for some answers to
94
+ questions you might have and/or if you want some other model quantized.
95
+
96
+ ## Thanks
97
+
98
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
99
+ me use its servers and providing upgrades to my workstation to enable
100
+ this work in my free time.
101
+
102
+ <!-- end -->