53482af554
feat(ui): add clipboard support to save/load preset
...
- add clipboard support to save/load preset with shift clicking
- update README.md for clarity
- fixes incorrect menu bar name for Load Preset
- update Czech translations
2024-12-17 21:05:58 -08:00
749f3215ec
feat(ui): support shift clicking to get quantization command
...
- support shift clicking Quantize Model button to get quantize command
- clean up imports in AutoGGUF.py and add localization keys
- use str() for getting log_dir_name
- remove legacy validate_quantization_inputs() function
- add return_command parameter to quantize_model() function
2024-11-12 19:41:59 -08:00
7575c97f6a
feat(config): support specifying log directory name
...
- support specifying log directory name using AUTOGGUF_LOG_DIR_NAME environment variable
- update dependencies, except transformers due to a small regression
2024-10-13 09:59:50 -07:00
35839eee77
refactor: use localizations for menubar and structure optimization
...
- use localizations for menubar
- bump AutoGGUF version to v2.0.0
- rename imports_and_globals.py to globals.py
- reformat code
- use file select for Merge/Split GGUF functions
- move general functions verify_gguf and process_args to globals.py
- create Plugins class for extensibility
2024-10-05 21:02:44 -07:00
b1b3a3549a
refactor: use correct strings for GGUF merging
2024-10-04 15:23:56 -07:00
e67992ec71
feat(gguf): add work in progress GGUF merge window
...
- localization strings and file select types incorrect
2024-09-29 23:42:04 -07:00
c831622d6b
feat(hf): add support for repository types
...
- add support for repository types in HF Transfer utility
- add dequantize_gguf.py script
- improve layout of HF Upload window
2024-09-22 09:48:48 -07:00
39441e503f
feat(convert): update llama.cpp convert scripts
...
- added support for MiniCPM3, RWKVv6, OLMoE, IBM Granite, and Jamba (conversion only: https://github.com/ggerganov/llama.cpp/pull/7531 )
- update gguf library from upstream
2024-09-20 15:05:52 -07:00
4e51ed2f56
refactor: update Finnish and Russian localizations
...
- update Finnish and Russian localizations using Claude 3 Opus
2024-09-17 17:41:50 -07:00
c02a02fbc1
feat(ui): update Japanese and German localizations
...
- update Japanese and German localizations
- bump version to v1.9.0
- optimize imports in main.py
- organize function definitions in AutoGGUF.py
2024-09-15 12:48:41 -07:00
1133422456
refactor: move functions out of AutoGGUF
...
- debloat AutoGGUF
- localize some GGUF split strings
2024-09-15 12:31:34 -07:00
79c74bb54a
feat(core): implement HF upload functionality
...
- add HF upload GUI definitions
- remove old HFTransfer class
- update localizations
2024-09-15 12:18:19 -07:00
747aa7b9a8
refactor: optimize GGUF imports
...
- optimize imports in GGUF conversion utilities
- rename gguf library modules
- update .gitignore and build workflow
2024-09-14 10:11:43 -07:00
3804da0a3f
feat(ui): add RAM and CPU usage graphs
...
- add RAM and CPU usage graphs
- add input validation using wraps
- reduce strictness of iMatrix status checking
- add right click context menu to models list
2024-09-10 15:58:17 -07:00
e5e18f9966
feat(core): add GGUF splitting feature
...
- add ability to split GGUFs
2024-09-09 20:46:53 -07:00
cee4294ecf
feat(parallel): add support for iMatrix generation tracking
...
- add support for iMatrix generation tracking
- don't adjust progress bar when indeterminate
2024-09-09 19:26:56 -07:00
fb092f8701
refactor: adjust logging and localizations
...
- adjust logging strings
- update French and Dutch localizations
- add minimal .env file loader in Localizations.py
2024-09-09 15:42:09 -07:00
be38e35d99
refactor: move functions into classes
...
- move functions into existing classes and files
- move AutoFP8 dialog out of a function and into __init__
2024-09-09 15:11:03 -07:00
e46c6260ee
fix: set progress to zero when indeterminate
2024-09-07 10:16:59 -07:00
454fb193f9
perf(startup): improve startup time by optimizing default configuration
...
- improve startup time by optimizing default configuration (disable network fetches for backends and updates)
- add startup time measuring
- use localizations for .env loader
2024-09-04 21:50:00 -07:00
8fb500a27d
refactor: remove requests and python-dotenv to reduce size
2024-09-04 21:18:49 -07:00
f8651ef19f
fix: reorganize imports and remove pytz
...
- reorganize imports and remove pytz from KVOverrideEntry.py
2024-09-04 17:53:32 -07:00
24ae0066ed
feat: support key shortcuts for AutoFP8 window
...
- update README.md for v1.8.1
- remove aliased quant types
- update .env.example with all configuration parameters
2024-09-04 17:31:39 -07:00
d55cb9ea9b
refactor: prepare repo for v1.8.1
...
- add localizations for AutoFP8
- remove unnecessary assets
- update CHANGELOG.md
- add DownloadThread.py docs
2024-09-04 17:19:54 -07:00
3adee266ca
feat(server): support API key authentication
...
- support authenticating via API key for the local server
- update Spanish localizations
2024-09-02 19:21:39 -07:00
a91f804ec1
fix: use proper status in TaskListItem
...
- use proper status in TaskListItem
- make sure to pass quant_threads and Logger to TaskListItem
- remove unnecessary logging in quantize_to_fp8_dynamic.py and optimize imports
2024-09-02 18:43:22 -07:00
a7f2dec852
feat(ui): add AutoFP8 quantization window
...
- add AutoFP8 quantization window (currently broken)
- add more dynamic KV parameters
2024-09-02 18:17:29 -07:00
79deff70ed
feat(window): add minimize/maximize button
...
- add minimize and maximize button to title bar
- improve smoothness of window moving
- prevent moving window below taskbar
2024-09-01 21:10:13 -07:00
81b2d4137b
feat(core): add AutoFP8 quantization classes
...
- add AutoFP8 quantization classes
- minor fixes based on IDE recommendations
2024-09-01 20:35:35 -07:00
22bd74b399
feat(server): replace Flask with FastAPI and Uvicorn
...
- replace Flask with FastAPI and Uvicorn
- fix web page not found error
- port is now defaulted to 7001
- bind to localhost (127.0.0.1) instead of 0.0.0.0
- improve performance by using Uvicorn
- add OpenAPI docs for endpoints
2024-08-31 15:34:47 -07:00
db1733b4ed
refactor: allow specifying output path in lora conversion subprocess
...
- remove the shutil.move operation
- allow specifying the output path in arguments in convert_lora_to_ggml.py utility
- bump max number of LoRA layers
2024-08-31 14:54:08 -07:00
a59b49fd97
chore: update llama.cpp convert scripts
2024-08-31 14:01:41 -07:00
fb9addb8c0
feat(models): add HF upload/download class
2024-08-31 13:54:24 -07:00
6583412b76
refactor: move functions out of AutoGGUF.py
...
- relocate functions into utils.py and TaskListItem.py
2024-08-29 15:01:13 -07:00
e307a4d3b5
chore: bump version to v1.8.0
2024-08-26 14:53:16 -07:00
fec22adc58
feat(config): add configuration options
...
- add more configuration options for AUTOGGUF_MODEL_DIR_NAME, AUTOGGUF_OUTPUT_DIR_NAME, and AUTOGGUF_RESIZE_FACTOR (these get created on startup)
- move some UI helper funtions out of AutoGGUF.py and into ui_update and Ta
- optimize imports for utility classes
- fix some missing imports
2024-08-25 15:11:36 -07:00
86d7bbba2d
style: reformat TaskListItem.py
2024-08-22 21:59:22 -07:00
ca9f9a4a76
fix: import missing modules
2024-08-22 21:59:01 -07:00
6e424462ab
refactor: add type hints
2024-08-22 21:56:37 -07:00
d4be39a22c
feat(core): implement plugins
...
- add plugins feature using importlib
- edit .gitignore
- change enabled state of AUTOGGUF_SERVER to "enabled" from "true" for consistency
2024-08-22 20:08:02 -07:00
53ab6a688e
fix: only count valid backends
...
- use valid_backends variable instead of backend_combo length, as the "No backends found" message is incorrectly counted as a single backend
2024-08-22 18:50:30 -07:00
e8026892e9
chore: fix Arabic and French localizations
2024-08-22 17:52:28 -07:00
a97a545a28
refactor: move get helper functions to utils.py
...
- move get_models_data and get_tasks_data to utils.py from AutoGGUF.py
2024-08-22 17:08:45 -07:00
4f2c8057e1
feat(core): add verification to manual import + concatenated file support
...
- verify GGUFs on manual import
- show warning when dealing with concatenated files such as mradermacher's split GGUFs (partXofX)
2024-08-22 15:57:21 -07:00
88875e3d67
feat(core): add manual model import
...
- allow importing models from any directory on the system
- add args as AutoGGUF class parameter
2024-08-22 15:39:08 -07:00
32f29b9f7c
fix: update regex for progress tracking
...
- update regex for progress tracking
- add .env.example file
- update .gitignore
2024-08-20 17:15:55 -07:00
c8babbd6e1
feat(core): add update checking support
...
- add update checking support with AUTOGGUF_CHECK_UPDATE (enabled) env variable
- fix setup.py
- add localization keys for update checking
- update version to v1.7.2 in Localizations.py
- formatting changes
2024-08-18 13:30:45 -07:00
a1650bba07
chore: add missing requirements
...
- add missing requirements
- add dotenv file loading
2024-08-17 13:27:18 -07:00
03c8a23e07
feat(ui): use Unicode for X button
...
- use Unicode for X button in KV Overrides box
- move CustomTitleBar to separate file
2024-08-17 13:24:37 -07:00
b62e8fc47c
feat(monitor): add smoother usage bar changes
2024-08-17 13:07:48 -07:00
BuildTools