Mirrored_Repos/InvokeAI
1
0
Fork 0
mirror of https://github.com/invoke-ai/InvokeAI synced 2024-08-30 20:32:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

docs: update CONFIGURATION.md

Browse Source 
Update model hashing docs
This commit is contained in:
psychedelicious 2024-03-15 00:09:34 +11:00
parent 328dc99f3a
commit bb8e6bbee6
1 changed files with 13 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
docs/features/CONFIGURATION.md
View File

@ -184,16 +184,26 @@ The provided token will be added as a `Bearer` token to the network requests to
### Model Hashing ### Model Hashing
Models are hashed during installation with the `BLAKE3` algorithm, providing a stable identifier for models across all platforms. Models are hashed during installation, providing a stable identifier for models across all platforms. The default algorithm is `blake3`, with a multi-threaded implementation.
Model hashing is a one-time operation, but it may take a couple minutes to hash a large model collection. You may opt out of model hashing and instead have a random UUID assigned instead: If your models are stored on a spinning hard drive, we suggest using `blake3_single`, the single-threaded implementation. The hashes are the same, but it's much faster on spinning disks.
```yaml ```yaml
InvokeAI: InvokeAI:
Model Install: Model Install:
skip_model_hash: true hashing_algorithm: blake3_single
``` ```
Model hashing is a one-time operation, but it may take a couple minutes to hash a large model collection. You may opt out of model hashing entirely by setting the algorithm to `random`.
```yaml
InvokeAI:
Model Install:
hashing_algorithm: random
```
Most common algorithms are supported, like `md5`, `sha256`, and `sha512`. These are typically much, much slower than `blake3`.
### Paths ### Paths
These options set the paths of various directories and files used by These options set the paths of various directories and files used by