Nguyen Ngoc Minh
dc4de43516
Merge pull request #6728 from menloresearch/fix/anthropic-model-load
...
Fix: Anthropic request to add models
2025-10-03 15:41:06 +00:00
Vanalite
9046927bf3
fix: optional custom header
2025-10-03 22:05:52 +07:00
Vanalite
7fd83e0bf8
feat: add migration for cohere provider
2025-10-03 21:55:28 +07:00
Faisal Amir
d3ea61b2e9
chore: update setting base url also
2025-10-03 21:49:42 +07:00
Faisal Amir
651ba446bf
chore: migration base_url anthropic
2025-10-03 21:40:59 +07:00
Vanalite
7bee282875
fix: Fix Cohere base url
2025-10-03 21:05:09 +07:00
Vanalite
86c7122f90
fix: Add header to anthropic get models api
2025-10-03 20:54:54 +07:00
Faisal Amir
c378d765fc
Merge pull request #6727 from menloresearch/fix/prompt-token
...
fix: prompt token
2025-10-03 20:12:12 +07:00
Faisal Amir
fdf239352a
chore: check model running
2025-10-03 19:52:15 +07:00
Faisal Amir
40c5953fea
fix:exclude enable thinking from FE
2025-10-03 19:06:01 +07:00
Faisal Amir
5382e9666e
fix: prompt token
2025-10-03 19:01:19 +07:00
Nguyen Ngoc Minh
404f40cc23
Merge pull request #6725 from menloresearch/ci/remove-upload-msi-step
...
ci: remove upload msi artifact
2025-10-03 06:13:22 +00:00
Minh141120
3ac83fdc08
ci: remove upload msi artifact
2025-10-03 13:11:45 +07:00
Nguyen Ngoc Minh
a568afdff1
Merge pull request #6721 from menloresearch/chore/use-custom-nsis-template
...
chore: use custom nsis template
2025-10-03 05:50:21 +00:00
Minh141120
14abd6068a
refacotr: clean up custom nsis template
2025-10-03 12:29:04 +07:00
Minh141120
d3a5ff9a74
chore: remove install windows deps from test
2025-10-03 12:06:57 +07:00
Minh141120
d7a1a9a529
feat: add logic for uninstall old jan version
2025-10-03 11:50:11 +07:00
Minh141120
4c4e795a3d
chore: use custom nsis template
2025-10-03 11:20:35 +07:00
Nghia Doan
c14e1ea00f
Merge pull request #6715 from menloresearch/fix/get-model-capabilities-correctly
...
fix: Extract model capabilities correctly for various providers on various platforms
2025-10-03 10:51:07 +07:00
Nguyen Ngoc Minh
f537429d2c
Merge pull request #6706 from menloresearch/qa/v0.7.0
...
feat: update checklist for 0.7.0
2025-10-02 08:30:30 +00:00
Minh141120
f6f9813ef2
feat: update checklist for 0.7.0
2025-10-02 15:26:37 +07:00
Nghia Doan
87db633b7d
Merge pull request #6700 from menloresearch/fix/edit-model-name
...
fix: Fix editing model without saving should restore original name
# Conflicts:
# web-app/src/containers/__tests__/EditModel.test.tsx
2025-10-02 09:28:50 +07:00
Nguyen Ngoc Minh
8e10f27cc2
Merge pull request #6701 from menloresearch/cherry-pick/projects
...
cherry pick : projects + performance enhancement
2025-10-01 16:42:20 +00:00
Dinh Long Nguyen
9f72debc17
fix: thread item overfetching ( #6699 )
...
* fix: thread item overfetching
* chore: cleanup left over import
2025-10-01 22:53:53 +07:00
Dinh Long Nguyen
1b9efee52c
feat: improve projects ( #6698 )
...
* decouple successfully
* only show movable projects for project items
* handle delete covnersations when projects is removed
* fix leftpanel assignemtn
* fix lint
2025-10-01 22:53:34 +07:00
Akarshan Biswas
0f0ba43b7f
feat: Adjust RAM/VRAM calculation for unified memory systems ( #6687 )
...
* feat: Adjust RAM/VRAM calculation for unified memory systems
This commit refactors the logic for calculating **total RAM** and **total VRAM** in `is_model_supported` and `plan_model_load` commands, specifically targeting systems with **unified memory** (like modern macOS devices where the GPU list may be empty).
The changes are as follows:
* **Total RAM Calculation:** If no GPUs are detected (`sys_info.gpus.is_empty()` is true), **total RAM** is now set to $0$. This avoids confusing total system memory with dedicated GPU memory when planning model placement.
* **Total VRAM Calculation:** If no GPUs are detected, **total VRAM** is still calculated as the system's **total memory (RAM)**, as this shared memory acts as VRAM on unified memory architectures.
This adjustment improves the accuracy of memory availability checks and model planning on unified memory systems.
* fix: total usable memory in case there is no system vram reported
* chore: temporarily change to self-hosted runner mac
* ci: revert back to github hosted runner macos
---------
Co-authored-by: Louis <louis@jan.ai>
Co-authored-by: Minh141120 <minh.itptit@gmail.com>
2025-10-01 18:58:14 +07:00
Nguyen Ngoc Minh
6a4aaaec87
Merge pull request #6694 from menloresearch/ci/revert-msi-installer
...
ci: revert upload msi to github release
2025-10-01 10:30:55 +00:00
Minh141120
a5574eaacb
ci: revert upload msi to github release
2025-10-01 17:00:03 +07:00
Faisal Amir
771d097309
Merge pull request #6688 from menloresearch/fix/dropdown-type-assistant
2025-10-01 14:47:54 +07:00
Louis
e0ab77cb24
fix: token count error ( #6680 )
2025-10-01 14:07:32 +07:00
Faisal Amir
7a36ed238c
Merge pull request #6681 from menloresearch/fix/local-api-server
...
fix: local api server auto start first model when last used missing
2025-10-01 14:01:22 +07:00
Faisal Amir
99d1713517
fix: dropdown type assistant
2025-10-01 14:00:55 +07:00
Faisal Amir
d102165028
chore: move auto start server setting
2025-10-01 11:44:49 +07:00
Faisal Amir
199623b414
chore: clear flow loacl api server
2025-10-01 11:23:59 +07:00
Faisal Amir
2679b19e32
fix: local api server auto start first model when missing last used
2025-10-01 11:04:28 +07:00
Nghia Doan
c5a5968bf8
Merge pull request #6643 from menloresearch/fix/model-name-change
...
fix: Apply model name change correctly
2025-09-30 22:41:05 +07:00
Nguyen Ngoc Minh
7eacee26d0
Merge pull request #6670 from menloresearch/ci/upload-msi-to-release
...
ci: add upload msi installer for windows
2025-09-30 21:15:34 +07:00
Minh141120
631a95e018
ci: add upload msi installer for windows
2025-09-30 20:12:53 +07:00
Nguyen Ngoc Minh
f7be59e99c
Merge pull request #6668 from menloresearch/feat/add-msi-installer
...
feat: add msi installer
2025-09-30 17:39:32 +07:00
Minh141120
508cbe16f8
refactor: remove redundant resource
2025-09-30 15:49:14 +07:00
Minh141120
dcb511023d
ci: add upload .msi artifact
2025-09-30 15:41:04 +07:00
Minh141120
0b8f3e01fb
feat: add msi installer for windows
2025-09-30 15:32:29 +07:00
Louis
de9c04474e
fix: should not allow duplicating model ( #6658 )
2025-09-30 13:59:17 +07:00
Faisal Amir
334b160012
Merge pull request #6656 from menloresearch/fix/thinking-block
...
fix: remove thinking tag on projects list message history
2025-09-30 13:48:34 +07:00
Nguyen Ngoc Minh
04fcd788a3
Merge pull request #6659 from menloresearch/fix/6626
...
fix: Improve KV cache estimation robustness
2025-09-30 13:42:28 +07:00
Akarshan
34b254e2d8
fix: Improve KV cache estimation robustness
...
The KV cache size calculation in estimate_kv_cache_internal now includes a fallback mechanism for models that do not explicitly define key_length and value_length in the GGUF metadata.
If these attention keys are missing, the head dimension (and thus key/value length) is calculated using the formula embedding_length / total_heads. This improves robustness and compatibility with GGUF models that don't have the proper keys in metadata.
Also adds logging of the full model metadata for easier debugging of the estimation process.
2025-09-30 11:14:18 +05:30
Nguyen Ngoc Minh
d315522c5a
Merge pull request #6618 from github-roushan/show-supported-files
...
Show supported files
2025-09-30 12:19:22 +07:00
Faisal Amir
1c9890649d
fix: remove thinking tag on projects list message history
2025-09-30 11:08:31 +07:00
Nghia Doan
3cbf348585
Merge pull request #6642 from menloresearch/fix/toast-text-selection-disable
...
feat: Disable text selection on Toaster
2025-09-30 11:01:31 +07:00
Louis
54d17c9c72
fix: migrate new mcp server config ( #6651 )
2025-09-30 00:07:57 +07:00
