Hugging Face's logo

mradermacher
/
mpt-30b-chat-i1-GGUF

Transformers
GGUF
English
Composer
MosaicML
llm-foundry
Inference Endpoints
imatrix
conversational
Model card Files Community
mpt-30b-chat-i1-GGUF / README.md
mradermacher's picture
mradermacher
auto-patch README.md
36fd686 verified
preview code
|
raw
history blame contribute delete
3.51 kB
metadata
base_model: mosaicml/mpt-30b-chat
datasets:
  - camel-ai/code
  - ehartford/wizard_vicuna_70k_unfiltered
  - anon8231489123/ShareGPT_Vicuna_unfiltered
  - timdettmers/openassistant-guanaco
  - camel-ai/math
  - camel-ai/biology
  - camel-ai/chemistry
  - camel-ai/ai_society
  - jondurbin/airoboros-gpt4-1.2
  - LongConversations
  - camel-ai/physics
language:
  - en
library_name: transformers
license: cc-by-nc-sa-4.0
no_imatrix: >-
  (IQ1_M) ggml_validate_row_data: found inf value at block 89
  llama_model_quantize: failed to quantize: quantized data validation failed
quantized_by: mradermacher
tags:
  - Composer
  - MosaicML
  - llm-foundry

About

weighted/imatrix quants of https://huggingface.co/mosaicml/mpt-30b-chat

*no more quants forthcoming, as llama.cpp corrupts them and crashes

static quants are available at https://huggingface.co/mradermacher/mpt-30b-chat-GGUF

Usage

If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs 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 i1-IQ2_M 10.2
GGUF i1-Q2_K 11.4 IQ3_XXS probably better
GGUF i1-IQ3_XXS 11.8 lower quality
GGUF i1-IQ3_M 14.6
GGUF i1-Q3_K_M 15.8 IQ3_S probably better
GGUF i1-Q4_K_S 17.2 optimal size/speed/quality
GGUF i1-Q4_K_M 19.2 fast, recommended
GGUF i1-Q6_K 24.7 practically like static Q6_K

Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):

image.png

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, for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to @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.