86 lines
5.9 KiB
Markdown
86 lines
5.9 KiB
Markdown
---
|
|
base_model: 922CA/Llama-3-monika-ddlc-11.5b-v1
|
|
datasets:
|
|
- 922-CA/MoCha_v1a
|
|
language:
|
|
- en
|
|
library_name: transformers
|
|
license: llama3
|
|
quantized_by: mradermacher
|
|
tags:
|
|
- text-generation-inference
|
|
- transformers
|
|
- unsloth
|
|
- llama
|
|
- trl
|
|
- sft
|
|
---
|
|
## About
|
|
|
|
<!-- ### quantize_version: 2 -->
|
|
<!-- ### output_tensor_quantised: 1 -->
|
|
<!-- ### convert_type: hf -->
|
|
<!-- ### vocab_type: -->
|
|
<!-- ### tags: nicoboss -->
|
|
weighted/imatrix quants of https://huggingface.co/922CA/Llama-3-monika-ddlc-11.5b-v1
|
|
|
|
<!-- provided-files -->
|
|
static quants are available at https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-GGUF
|
|
## Usage
|
|
|
|
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
|
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
|
|
more details, including on how to concatenate multi-part files.
|
|
|
|
## Provided Quants
|
|
|
|
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
|
|
|
|
| Link | Type | Size/GB | Notes |
|
|
|:-----|:-----|--------:|:------|
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ1_S.gguf) | i1-IQ1_S | 2.9 | for the desperate |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ1_M.gguf) | i1-IQ1_M | 3.1 | mostly desperate |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 3.4 | |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ2_XS.gguf) | i1-IQ2_XS | 3.7 | |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ2_S.gguf) | i1-IQ2_S | 3.9 | |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ2_M.gguf) | i1-IQ2_M | 4.2 | |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q2_K_S.gguf) | i1-Q2_K_S | 4.3 | very low quality |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q2_K.gguf) | i1-Q2_K | 4.6 | IQ3_XXS probably better |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 4.7 | lower quality |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ3_XS.gguf) | i1-IQ3_XS | 5.0 | |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q3_K_S.gguf) | i1-Q3_K_S | 5.3 | IQ3_XS probably better |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ3_S.gguf) | i1-IQ3_S | 5.3 | beats Q3_K* |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ3_M.gguf) | i1-IQ3_M | 5.4 | |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q3_K_M.gguf) | i1-Q3_K_M | 5.8 | IQ3_S probably better |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q3_K_L.gguf) | i1-Q3_K_L | 6.3 | IQ3_M probably better |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ4_XS.gguf) | i1-IQ4_XS | 6.4 | |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q4_0.gguf) | i1-Q4_0 | 6.7 | fast, low quality |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-IQ4_NL.gguf) | i1-IQ4_NL | 6.7 | prefer IQ4_XS |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q4_K_S.gguf) | i1-Q4_K_S | 6.8 | optimal size/speed/quality |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q4_K_M.gguf) | i1-Q4_K_M | 7.1 | fast, recommended |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q4_1.gguf) | i1-Q4_1 | 7.4 | |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q5_K_S.gguf) | i1-Q5_K_S | 8.1 | |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q5_K_M.gguf) | i1-Q5_K_M | 8.3 | |
|
|
| [GGUF](https://huggingface.co/mradermacher/Llama-3-monika-ddlc-11.5b-v1-i1-GGUF/resolve/main/Llama-3-monika-ddlc-11.5b-v1.i1-Q6_K.gguf) | i1-Q6_K | 9.6 | practically like static Q6_K |
|
|
|
|
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
|
types (lower is better):
|
|
|
|
|
|
|
|
And here are Artefact2's thoughts on the matter:
|
|
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
|
|
|
|
## FAQ / Model Request
|
|
|
|
See https://huggingface.co/mradermacher/model_requests for some answers to
|
|
questions you might have and/or if you want some other model quantized.
|
|
|
|
## Thanks
|
|
|
|
I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
|
|
me use its servers and providing upgrades to my workstation to enable
|
|
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.
|
|
|
|
<!-- end -->
|