mradermacher's picture
auto-patch README.md
2db2c7b verified
|
raw
history blame
3.63 kB
---
exported_from: TigerResearch/tigerbot-70b-chat-v4
language:
- en
library_name: transformers
license: apache-2.0
quantized_by: mradermacher
---
## About
<!-- ### quantize_version: 1 -->
<!-- ### output_tensor_quantised: 1 -->
<!-- ### convert_type: -->
<!-- ### vocab_type: -->
weighted/imatrix quants of https://huggingface.co/TigerResearch/tigerbot-70b-chat-v4
<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-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/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-IQ2_M.gguf) | i1-IQ2_M | 23.6 | |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q2_K.gguf) | i1-Q2_K | 25.9 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 27.0 | lower quality |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q3_K_S.gguf) | i1-Q3_K_S | 30.4 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q3_K_M.gguf) | i1-Q3_K_M | 33.7 | IQ3_S probably better |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q3_K_L.gguf) | i1-Q3_K_L | 36.6 | IQ3_M probably better |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-IQ4_XS.gguf) | i1-IQ4_XS | 37.3 | |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q4_0.gguf) | i1-Q4_0 | 39.5 | fast, low quality |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q4_K_S.gguf) | i1-Q4_K_S | 39.7 | optimal size/speed/quality |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q4_K_M.gguf) | i1-Q4_K_M | 41.9 | fast, recommended |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q5_K_S.gguf) | i1-Q5_K_S | 48.0 | |
| [GGUF](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q5_K_M.gguf) | i1-Q5_K_M | 49.3 | |
| [PART 1](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/tigerbot-70b-chat-v4-i1-GGUF/resolve/main/tigerbot-70b-chat-v4.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 57.1 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):
![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
## 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.
<!-- end -->