mirrors/llama.vscode
2
0
Fork 0
mirror of https://github.com/ggml-org/llama.vscode.git synced 2026-05-07 01:15:23 +00:00
Code Issues Projects Releases Packages Wiki Activity
157 commits 52 branches 47 tags 5.1 MiB
Commit graph

157 commits

This branch
This branch
All branches
Author SHA1 Message Date
igardev
79b90ef999
 		v0.0.47 v0.0.47 2026-05-04 08:42:41 +03:00
igardev
a544d93511
 		API key for getting models list, ultiline field for Edit with AI, qwen3.5 models added 
- API key is used (if needed and provided) on getting the list of models for adding OpenAI compatible provider
- Multiline field for Edit with AI
- Qwen3.5 models (2B, 4B, 9B) added in the predefined list  - good for tools and chat
 2026-05-04 08:42:01 +03:00
Alexey Mekhanoshin
a73d9498ab
 		feat: add authorization headers to models fetch request (#180) 
Adds support for the Authorization header when fetching the list of models from an OpenAI-compatible provider.
 2026-04-30 07:59:35 +03:00
igardev
29f6c9973b
 		v0.0.46 v0.0.46 2026-04-29 21:06:11 +03:00
Copilot
f98919badf
 		Add llama.vscode model provider for GitHub Copilot Chat (#171) 
With this change llama.vscode could provide models for VS Code Copilot:
1. Start tools model from llama-vscode (local or external)  
2. In VS Code Copilot show the models list -> Other Models -> Manage Models  
3. Make the models (all models available by the application serving the tools model are shown) you want to use visible (click on the left of the model name)  
4. Select the desired model from Copilot and start using it

Not needed tools from Copilot could be unchecked to reduce contex size if local model is used.
 2026-04-29 21:04:33 +03:00
igardev
63c25e6dc9
 		v0.0.45 v0.0.45 2026-03-04 08:23:13 +02:00
Haafiz
0a588177b7
 		feat: add configurable debounce for inline completion requests (#164) 
Waits for the user to pause typing before sending a request to the server.
Set to 0 (default) to disable.
 2026-03-04 08:18:42 +02:00
Haafiz
caa0f9363d
 		fix: show 'extension is updated' notification only on actual version change (#167) 
Add version tracking and update persistence keys
 2026-03-04 08:06:32 +02:00
igardev
f8158d9e48
 		v0.0.44 v0.0.44 2026-03-02 23:53:18 +02:00
igardev
bf7d0c2892
 		Subagents (#169) 
* Read SOUL.md and USER.md files from project root and add them in the prompt if they exist (similar to OpenClaw).

* - Subagents implemented
- new agent Unit Test Writer
- new tool create_agent
- new agent "Agent creator"

* Update documentation for llama-vscode
 2026-03-02 23:52:38 +02:00
igardev
15539816d3
 		v0.0.43 v0.0.43 2026-02-18 17:32:31 +02:00
Simon Vogl
9a22c2019b
 		Fix typo in label from 'Descriptoin' to 'Description' (#160) 
minimal typo fixed
 2026-02-18 06:42:46 +02:00
Manzoor Ahmed Shaikh
fc08459518
 		Typo of 'description' corrected, closes #162 (#163) 2026-02-18 06:41:10 +02:00
crsawyer
3cef9a4947
 		Add AGENTS.md loading from workspace root (#158) 
- Only if `llama-vscode-rules.md` is not present to preserve existing functionality
- TODO: Implement nested AGENTS.md
- See: https://agents.md/

Co-authored-by: Caleb Sawyer <caleb.sawyer@gsdeng.com>
 2026-02-18 06:40:16 +02:00
Georgi Gerganov
0182f4fa00
 		Use fixed slot id for FIM requests (#155) 2026-02-08 23:31:13 +02:00
igardev
3ba0632316
 		v0.0.42 v0.0.42 2026-01-18 22:03:29 +02:00
igardev
dbdc8f523d
 		Add health check for models (works only with llama.cpp server) (#156) 
- Add health check for models (works only with llama.cpp server)
- Visible in environment view
New settings:
- Health_check_interval_s: The interval in seconds for the health check
- Health_check_compl_enabled: Enable/disable health check for completion model
- Health_check_chat_enabled: Enable/disable health check for chat model
- Health_check_embs_enabled: Enable/disable health check for embedding model
- Health_check_tools_enabled: Enable/disable health check for tools model
 2026-01-18 22:02:40 +02:00
igardev
9d48f91b19
 		v0.0.41 v0.0.41 2026-01-07 18:45:31 +02:00
igardev
a035f96de4
 		Fix code completion in case max_parallel_completions = 1 or empty. (#150) 2026-01-07 18:44:35 +02:00
igardev
bdbe2e31d5
 		v0.0.40 v0.0.40 2026-01-05 15:44:56 +02:00
igardev
8e42441edc
 		Generate multiple completions in parallel (#148) 
- Setting max_parallel_completions determines how many completions to generate in parallel (default 3)
- Shortcuts - Alt+] - next completion, Alt+[ - previous completion
- Requires llama.cpp after December, 6, 2025 (commit c42712b) but is backword compatible (generates one completion for older versions)
 2026-01-05 15:43:33 +02:00
igardev
7fb8190433
 		v0.0.39 v0.0.39 2025-12-31 17:38:16 +02:00
igardev
1d5f9387ab
 		Enable skills usage (#147) 
- Skills (https://agentskills.io/home) could be now parsed by the LLM and added in the prompt
- skills_folder setting determines where are skills descriptions. If empty the <project_folder>/skills folder is used by default
- Anthropic models support skills best. I guess, the open source models will catch up.
 2025-12-31 17:36:24 +02:00
igardev
b8d0e1af8d
 		v0.0.38 v0.0.38 2025-12-30 19:37:13 +02:00
igardev
ed3270d08a
 		Agent todo tool (#146) 
- update_todo_list  tool added - now the agent could plan and track the execution of user request by todo items
 2025-12-30 19:35:57 +02:00
igardev
28519f8c66
 		v0.0.37 v0.0.37 2025-11-18 17:38:32 +02:00
igardev
e5a56f269d
 		Image selection for agent context (#140) 
* Image selection is now possible in Agent view

* Add qwen3 VL 30B Instruct as a predefined model
 2025-11-18 17:37:01 +02:00
igardev
69f059e7aa
 		Reduce the size of the installation package (#135) v0.0.36 2025-11-04 11:02:04 +02:00
igardev
99e85287ce
 		v0.0.36 2025-11-03 00:14:00 +02:00
igardev
c03e32623e
 		Adding a default model for agent is not possible (#134) 
Now the agent could have a default tools model. If there is such a model, on selecting the agent, the (tools) agent model will also be selected. However, it is still possible to change the tools model, while using the same agent.
 2025-11-03 00:12:48 +02:00
Nuno
723432b54a
 		fix: use correct function for last used env on startup (#133) 
Signed-off-by: rare-magma <rare-magma@posteo.eu>
 2025-11-02 23:49:21 +02:00
Georgi Gerganov
f610bd28e7
 		readme : add Qwen3 30B 2025-11-01 18:49:10 +02:00
igardev
7a57e41c05 test 2025-10-31 18:22:32 +02:00
igardev
bbba713e02
 		v0.0.35 v0.0.35 2025-10-27 02:49:14 +02:00
igardev
73eaa326e6
 		- Show error in agent view UI in case of error on editing file (#131) 
- Show error in agent view UI in case of error on editing file
- Other bug fixes
 2025-10-27 02:47:50 +02:00
stoperro
5e565da06b
 		Working multi-file edit PoC. (#130) 2025-10-25 21:34:08 +03:00
igardev
fab5072a1e
 		v0.0.34 v0.0.34 2025-10-14 14:49:11 +03:00
igardev
b91585925d
 		Edit Agent View (#128) 
- Edit Agent view added
- Add model from OpenAI compatible provider
 2025-10-14 14:48:28 +03:00
igardev
889c5f6d57
 		Remove the todo from the agent view (#127) v0.0.33 
Co-authored-by: igardev <ivailo.gardev@akros.ch>
 2025-10-08 17:44:06 +03:00
igardev
cfcdfb0288
 		v0.0.33 2025-10-08 17:34:28 +03:00
igardev
43416b7d3c
 		Bug fixes (#126) 
- Fix the embeddings ranking - the wrong query was used
- Removed the free DeepSeek model from open router as not working now
- Setting endpoint_tools is now enought to start agent
- Agent View is now not hidden if the tools model is deselected
- Refactoring
 2025-10-08 16:59:23 +03:00
igardev
ff415359dd
 		Reduce the size of the extension (#125) v0.0.32 2025-10-05 16:05:17 +03:00
igardev
a5089febb9
 		v0.0.32 2025-10-05 15:47:07 +03:00
igardev
13bf27699e
 		Menu refactor and Ask agent added (#124) 
* menu.ts is refactored - services classes are extracted

* - Refactor menu.ts model - extract services
- agent "Ask" added for questions about the project without changing the files
- predefiled free models from OpenRouter added (and xAi removed as not free anymore)
- Some bugs fixed

---------

Co-authored-by: igardev <ivailo.gardev@akros.ch>
 2025-10-05 15:44:04 +03:00
igardev
7195812083
 		v0.0.31 v0.0.31 2025-10-03 00:50:52 +03:00
igardev
fa9aa34755
 		Refactor menu.ts and bug fixes 
- Fix bug for adding files by the agent
- Extract agent service from menu
- Adding agent from menu is now possible (not very user friendly, but working)
 2025-10-03 00:50:10 +03:00
igardev
74a24f8a7e
 		v0.0.30 v0.0.30 2025-10-02 14:23:49 +03:00
igardev
8b0943ec8f
 		Bug fixes 
- launch commands fixed (some chars were escaped)
- Agent chat is visualized on opening chat
 2025-10-02 14:23:14 +03:00
igardev
91d7414aa0
 		v0.0.29 v0.0.29 2025-10-01 10:04:49 +03:00
igardev
fd964e708e
 		- Menu.ts refactored (#119) 
- Bug with editing files by agent fixed
- Menu.ts refactored
 2025-10-01 10:04:01 +03:00