Hugging Face's logo

RichardErkhov
/
Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf

GGUF
Inference Endpoints
Model card Files Community
RichardErkhov commited on
Commit
fb22d6b
•
1 Parent(s): 3cd26b6

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +92 -0
README.md ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Quantization made by Richard Erkhov.
2
+
3
+ [Github](https://github.com/RichardErkhov)
4
+
5
+ [Discord](https://discord.gg/pvy7H8DZMG)
6
+
7
+ [Request more models](https://github.com/RichardErkhov/quant_request)
8
+
9
+
10
+ TinyLlama-1.1b-karasu-merged - GGUF
11
+ - Model creator: https://huggingface.co/Kame1024/
12
+ - Original model: https://huggingface.co/Kame1024/TinyLlama-1.1b-karasu-merged/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [TinyLlama-1.1b-karasu-merged.Q2_K.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q2_K.gguf) | Q2_K | 0.4GB |
18
+ | [TinyLlama-1.1b-karasu-merged.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.IQ3_XS.gguf) | IQ3_XS | 0.44GB |
19
+ | [TinyLlama-1.1b-karasu-merged.IQ3_S.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.IQ3_S.gguf) | IQ3_S | 0.47GB |
20
+ | [TinyLlama-1.1b-karasu-merged.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q3_K_S.gguf) | Q3_K_S | 0.47GB |
21
+ | [TinyLlama-1.1b-karasu-merged.IQ3_M.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.IQ3_M.gguf) | IQ3_M | 0.48GB |
22
+ | [TinyLlama-1.1b-karasu-merged.Q3_K.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q3_K.gguf) | Q3_K | 0.51GB |
23
+ | [TinyLlama-1.1b-karasu-merged.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q3_K_M.gguf) | Q3_K_M | 0.51GB |
24
+ | [TinyLlama-1.1b-karasu-merged.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q3_K_L.gguf) | Q3_K_L | 0.55GB |
25
+ | [TinyLlama-1.1b-karasu-merged.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.IQ4_XS.gguf) | IQ4_XS | 0.57GB |
26
+ | [TinyLlama-1.1b-karasu-merged.Q4_0.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q4_0.gguf) | Q4_0 | 0.59GB |
27
+ | [TinyLlama-1.1b-karasu-merged.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.IQ4_NL.gguf) | IQ4_NL | 0.6GB |
28
+ | [TinyLlama-1.1b-karasu-merged.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q4_K_S.gguf) | Q4_K_S | 0.6GB |
29
+ | [TinyLlama-1.1b-karasu-merged.Q4_K.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q4_K.gguf) | Q4_K | 0.62GB |
30
+ | [TinyLlama-1.1b-karasu-merged.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q4_K_M.gguf) | Q4_K_M | 0.62GB |
31
+ | [TinyLlama-1.1b-karasu-merged.Q4_1.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q4_1.gguf) | Q4_1 | 0.65GB |
32
+ | [TinyLlama-1.1b-karasu-merged.Q5_0.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q5_0.gguf) | Q5_0 | 0.71GB |
33
+ | [TinyLlama-1.1b-karasu-merged.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q5_K_S.gguf) | Q5_K_S | 0.71GB |
34
+ | [TinyLlama-1.1b-karasu-merged.Q5_K.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q5_K.gguf) | Q5_K | 0.73GB |
35
+ | [TinyLlama-1.1b-karasu-merged.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q5_K_M.gguf) | Q5_K_M | 0.73GB |
36
+ | [TinyLlama-1.1b-karasu-merged.Q5_1.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q5_1.gguf) | Q5_1 | 0.77GB |
37
+ | [TinyLlama-1.1b-karasu-merged.Q6_K.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q6_K.gguf) | Q6_K | 0.84GB |
38
+ | [TinyLlama-1.1b-karasu-merged.Q8_0.gguf](https://huggingface.co/RichardErkhov/Kame1024_-_TinyLlama-1.1b-karasu-merged-gguf/blob/main/TinyLlama-1.1b-karasu-merged.Q8_0.gguf) | Q8_0 | 1.09GB |
39
+
40
+
41
+
42
+
43
+ Original model description:
44
+ ---
45
+ base_model:
46
+ - lightblue/karasu-1.1B
47
+ - TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
48
+ - TinyLlama/TinyLlama-1.1B-Chat-v1.0
49
+ library_name: transformers
50
+ tags:
51
+ - mergekit
52
+ - merge
53
+
54
+ ---
55
+ # storage
56
+
57
+ This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
58
+
59
+ ## Merge Details
60
+ ### Merge Method
61
+
62
+ This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) as a base.
63
+
64
+ ### Models Merged
65
+
66
+ The following models were included in the merge:
67
+ * [lightblue/karasu-1.1B](https://huggingface.co/lightblue/karasu-1.1B)
68
+ * [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0)
69
+
70
+ ### Configuration
71
+
72
+ The following YAML configuration was used to produce this model:
73
+
74
+ ```yaml
75
+
76
+ models:
77
+ - model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
78
+ parameters:
79
+ weight: 0.5
80
+ - model: lightblue/karasu-1.1B
81
+ parameters:
82
+ weight: 0.5
83
+ merge_method: dare_ties
84
+ base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
85
+ parameters:
86
+ normalize: true
87
+ int8_mask: true
88
+ dtype: bfloat16
89
+
90
+ ```
91
+
92
+