Downloads · 30 days
9
6% of all-time downloads
RichardErkhov/llmixer_-_BigWeave-v27-95b-gguf
llmixer_-_BigWeave-v27-95b-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
9
6% of all-time downloads
All-time downloads
142
Public
Repo size
1.1 TB
Likes
0
Public
Click a slice to open those files.
.gguf1.3 TB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
BigWeave-v27-95b - GGUF
| Name | Quant method | Size |
|---|---|---|
| BigWeave-v27-95b.Q2_K.gguf | Q2_K | 32.98GB |
| BigWeave-v27-95b.IQ3_XS.gguf | IQ3_XS | 36.69GB |
| BigWeave-v27-95b.IQ3_S.gguf | IQ3_S | 38.78GB |
| BigWeave-v27-95b.Q3_K_S.gguf | Q3_K_S | 38.66GB |
| BigWeave-v27-95b.IQ3_M.gguf | IQ3_M | 40.12GB |
| BigWeave-v27-95b.Q3_K.gguf | Q3_K | 43.16GB |
| BigWeave-v27-95b.Q3_K_M.gguf | Q3_K_M | 43.16GB |
| BigWeave-v27-95b.Q3_K_L.gguf | Q3_K_L | 47.01GB |
| BigWeave-v27-95b.IQ4_XS.gguf | IQ4_XS | 48.37GB |
| BigWeave-v27-95b.Q4_0.gguf | Q4_0 | 50.55GB |
| BigWeave-v27-95b.IQ4_NL.gguf | IQ4_NL | 51.04GB |
| BigWeave-v27-95b.Q4_K_S.gguf | Q4_K_S | 50.94GB |
| BigWeave-v27-95b.Q4_K.gguf | Q4_K | 53.82GB |
| BigWeave-v27-95b.Q4_K_M.gguf | Q4_K_M | 53.82GB |
| BigWeave-v27-95b.Q4_1.gguf | Q4_1 | 56.14GB |
| BigWeave-v27-95b.Q5_0.gguf | Q5_0 | 61.74GB |
| BigWeave-v27-95b.Q5_K_S.gguf | Q5_K_S | 61.74GB |
| BigWeave-v27-95b.Q5_K.gguf | Q5_K | 63.42GB |
| BigWeave-v27-95b.Q5_K_M.gguf | Q5_K_M | 63.42GB |
| BigWeave-v27-95b.Q5_1.gguf | Q5_1 | 67.33GB |
| BigWeave-v27-95b.Q6_K.gguf | Q6_K | 73.62GB |
| BigWeave-v27-95b.Q8_0.gguf | Q8_0 | 95.35GB |
base_model:
The BigWeave models aim to experimentally identify merge settings for increasing model performance. The version number merely tracks various attempts and is not a quality indicator. Only results demonstrating good performance are retained and shared.
Chatml, Mistral, Vicuna.
This is a self-merge of 152334H/miqu-1-70b-sf. The 30 most important layers (according to exl2 measurements) are duplicated with 50% overlap.
Merge configuration:
slices:
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [0,40]
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [34,45] # dup 34-44
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [40,52]
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [51,53] # dup 51-52
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [52,55]
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [54,56] # dup 54-55
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [55,59]
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [58,60] # dup 58-59
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [59,72]
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [64,79] # dup 64-78
- sources:
- model: 152334H/miqu-1-70b-sf
layer_range: [72,80]
merge_method: passthrough
dtype: float16