Commit graph

4807 commits

Author SHA1 Message Date
oobabooga 03fb85e49a Update llama.cpp 2025-07-15 07:37:13 -07:00
oobabooga 845432b9b4 Remove the obsolete modules/relative_imports.py file 2025-07-14 21:03:18 -07:00
oobabooga 1d1b20bd77 Remove the --torch-compile option (it doesn't do anything currently) 2025-07-11 10:51:23 -07:00
oobabooga 5a8a9c22e8 Update llama.cpp 2025-07-11 09:20:27 -07:00
oobabooga 273888f218 Revert "Use eager attention by default instead of sdpa"
This reverts commit bd4881c4dc.
2025-07-10 18:56:46 -07:00
oobabooga caf69d871a Revert "Standardize margins and paddings across all chat styles"
This reverts commit 86cb5e0587.
2025-07-10 18:43:01 -07:00
oobabooga 188c7c8f2b Revert "CSS simplifications"
This reverts commit c6c1b725e9.
2025-07-10 18:42:52 -07:00
oobabooga 635e6efd18 Ignore add_bos_token in instruct prompts, let the jinja2 template decide 2025-07-10 07:14:01 -07:00
oobabooga 0f3a88057c Don't downgrade triton-windows on CUDA 12.8 2025-07-10 05:39:04 -07:00
oobabooga e523f25b9f Downgrade triton-windows to 3.2.0.post19
https://github.com/oobabooga/text-generation-webui/issues/7107#issuecomment-3057250374
2025-07-10 05:35:57 -07:00
oobabooga a7a3a0c700 Merge remote-tracking branch 'refs/remotes/origin/dev' into dev 2025-07-09 21:07:42 -07:00
oobabooga 21e0e9f32b Add the triton-windows requirement on Windows to make transformers functional 2025-07-09 21:05:17 -07:00
dependabot[bot] d1f4622a96
Update peft requirement from ==0.15.* to ==0.16.* in /requirements/full (#7127) 2025-07-10 00:15:50 -03:00
oobabooga e015355e4a Update README 2025-07-09 20:03:53 -07:00
oobabooga bd4881c4dc Use eager attention by default instead of sdpa 2025-07-09 19:57:37 -07:00
oobabooga b69f435311 Fix latest transformers being super slow 2025-07-09 19:56:50 -07:00
oobabooga 8b3c7aa795 Bump bitsandbytes to 0.46 2025-07-09 19:46:55 -07:00
oobabooga f045b72826 Bump accelerate to 1.8 2025-07-09 19:46:26 -07:00
oobabooga c357601c01 Bump transformers to 4.53 2025-07-09 18:48:04 -07:00
oobabooga 6c2bdda0f0 Transformers loader: replace use_flash_attention_2/use_eager_attention with a unified attn_implementation
Closes #7107
2025-07-09 18:39:37 -07:00
oobabooga 511bb31646 Merge remote-tracking branch 'refs/remotes/origin/dev' into dev 2025-07-08 20:04:37 -07:00
oobabooga d1e9301a43 Remove fragile js from 9a58964834 2025-07-08 19:57:46 -07:00
Cats cd5d867b62
docs: Add Mirostat Explanation (#7128) 2025-07-08 17:54:38 -03:00
oobabooga 3e24a127c7 Remove more unnecessary files from portable builds 2025-07-08 09:13:11 -07:00
oobabooga 2f544fe199 Update the keyboard shortcuts documentation 2025-07-08 09:02:42 -07:00
oobabooga 93e08c0d4a Update README 2025-07-08 08:59:29 -07:00
oobabooga 42191a36ab Keep navigation icons visible when switching versions 2025-07-08 07:10:04 -07:00
oobabooga c6c1b725e9 CSS simplifications 2025-07-07 21:11:13 -07:00
oobabooga 86cb5e0587 Standardize margins and paddings across all chat styles 2025-07-07 21:02:19 -07:00
oobabooga e8266b0356 Use windows-2022 in workflows 2025-07-07 14:19:20 -07:00
oobabooga 74d98186fc Slightly more robust autoscroll 2025-07-07 13:23:23 -07:00
oobabooga ca226a54c6 Disable the message version navigation hover effects during streaming 2025-07-07 11:29:37 -07:00
oobabooga 07e6f004c5 Rename a button in the Session tab for clarity 2025-07-07 11:28:47 -07:00
oobabooga 426e7a4cec Update the extensions documentation 2025-07-07 08:43:01 -07:00
oobabooga e52bc0acb2 Update llama.cpp 2025-07-06 20:28:35 -07:00
oobabooga cbef2720ce Revert "Fix: use embedded Python in start_windows.bat to avoid system interpreter conflicts (#7120)"
This reverts commit 8df1127ce2.
2025-07-06 20:14:02 -07:00
Alidr79 e5767d4fc5
Update ui_model_menu.py blocking the --multi-user access in backend (#7098) 2025-07-06 21:48:53 -03:00
oobabooga 60123a67ac Better log message when extension requirements are not found 2025-07-06 17:44:41 -07:00
oobabooga e6bc7742fb Support installing user extensions in user_data/extensions/ 2025-07-06 17:30:23 -07:00
Philipp Claßen 959d4ddb91
Fix for chat sidebars toggle buttons disappearing (#7106) 2025-07-06 20:51:42 -03:00
Ali 8df1127ce2
Fix: use embedded Python in start_windows.bat to avoid system interpreter conflicts (#7120) 2025-07-06 20:42:34 -03:00
oobabooga de4ccffff8 Fix the duckduckgo search 2025-07-06 16:24:57 -07:00
oobabooga 0f258774d3 Minor README changes 2025-07-05 14:25:59 -07:00
oobabooga 4583924ce7 Remove torchvision/torchaudio mentions from the README 2025-07-05 14:24:15 -07:00
oobabooga c4d738f39f Update llama.cpp 2025-07-05 14:09:29 -07:00
oobabooga c4d5331c03 Fix autoscroll after fonts load 2025-07-04 13:21:52 -07:00
oobabooga 92ec8dda03 Fix chat history getting lost if the UI is inactive for a long time (closes #7109) 2025-07-04 06:04:04 -07:00
oobabooga 23bb94a5fb Update llama.cpp 2025-07-03 20:36:54 -07:00
zombiegreedo 877c651c04
Handle either missing <think> start or </think> end tags (#7102) 2025-07-03 23:05:46 -03:00
oobabooga cbba88f565 Fix scrolling during streaming when thinking blocks are present 2025-07-03 18:16:29 -07:00