Downloads · 30 days
130
6% of all-time downloads
RichardErkhov/ddh0_-_OrcaMaid-v2-FIX-13b-32k-gguf
ddh0_-_OrcaMaid-v2-FIX-13b-32k-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
130
6% of all-time downloads
All-time downloads
2.1K
Public
Repo size
116 GB
Likes
0
Public
Click a slice to open those files.
.gguf133 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
OrcaMaid-v2-FIX-13b-32k - GGUF
| Name | Quant method | Size |
|---|---|---|
| OrcaMaid-v2-FIX-13b-32k.Q2_K.gguf | Q2_K | 4.52GB |
| OrcaMaid-v2-FIX-13b-32k.Q3_K_S.gguf | Q3_K_S | 5.27GB |
| OrcaMaid-v2-FIX-13b-32k.Q3_K.gguf | Q3_K | 3.0GB |
| OrcaMaid-v2-FIX-13b-32k.Q3_K_M.gguf | Q3_K_M | 5.9GB |
| OrcaMaid-v2-FIX-13b-32k.Q3_K_L.gguf | Q3_K_L | 3.76GB |
| OrcaMaid-v2-FIX-13b-32k.IQ4_XS.gguf | IQ4_XS | 6.54GB |
| OrcaMaid-v2-FIX-13b-32k.Q4_0.gguf | Q4_0 | 1.95GB |
| OrcaMaid-v2-FIX-13b-32k.IQ4_NL.gguf | IQ4_NL | 3.08GB |
| OrcaMaid-v2-FIX-13b-32k.Q4_K_S.gguf | Q4_K_S | 6.88GB |
| OrcaMaid-v2-FIX-13b-32k.Q4_K.gguf | Q4_K | 7.33GB |
| OrcaMaid-v2-FIX-13b-32k.Q4_K_M.gguf | Q4_K_M | 7.33GB |
| OrcaMaid-v2-FIX-13b-32k.Q4_1.gguf | Q4_1 | 7.61GB |
| OrcaMaid-v2-FIX-13b-32k.Q5_0.gguf | Q5_0 | 8.36GB |
| OrcaMaid-v2-FIX-13b-32k.Q5_K_S.gguf | Q5_K_S | 3.06GB |
| OrcaMaid-v2-FIX-13b-32k.Q5_K.gguf | Q5_K | 8.6GB |
| OrcaMaid-v2-FIX-13b-32k.Q5_K_M.gguf | Q5_K_M | 8.6GB |
| OrcaMaid-v2-FIX-13b-32k.Q5_1.gguf | Q5_1 | 9.1GB |
| OrcaMaid-v2-FIX-13b-32k.Q6_K.gguf | Q6_K | 9.95GB |
| OrcaMaid-v2-FIX-13b-32k.Q8_0.gguf | Q8_0 | 12.88GB |
This is the fixed version of OrcaMaid-v2-13b, extended to 32768 context length via YaRN. The (now-deleted) v2 model had issues with the merged tokenizer that prevented it from stopping when necessary, and caused it to output broken ChatML tokens like <|im_end, etc.
This is a gradient SLERP merge of Microsoft's Orca-2-13b and Undi and IkariDev's Noromaid-v0.1.1-13b, biased towards Orca.
Just as with OrcaMaid v1, the overall goal of this merge is to create a model that sounds uniquely human and natural, without sacrificing intelligence.
The prompt format is Alpaca. You can use the standard format as shown, but for best results, you should customize the system prompt to your specific needs.
Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{YOUR MESSAGE HERE}
### Response:
{BOT MESSAGE HERE}
<s></s>32768 via YaRN (original context length was 4096)