Ling0925
/
suna
mirror of https://github.com/kortix-ai/suna.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
2,475 Commits 30 Branches 7 Tags 406 MiB
Commit Graph

42 Commits

Author SHA1 Message Date
Saumya 7e10d736b5 chore: cleanup llm modules 2025-08-22 16:45:26 +05:30
marko-kraemer 535fc03031 default model 2025-08-21 21:23:43 -07:00
marko-kraemer 4cacf55c5b Merge branch 'main' into config-refactor 2025-08-17 11:45:09 -07:00
Krishav Raj Singh 4838b99c24 refactor + add num_retries 2025-08-17 23:26:11 +05:30
marko-kraemer 1896208f12 refactor wip 2025-08-16 19:10:56 -07:00
Vukasin e779ae0fe4 feat: playbook fixes and unified modal 2025-08-10 17:41:36 +02:00
Vukasin 4d2d787974 fix(llm): enable litellm.drop_params and map GPT-5 max_tokens to max_completion_tokens; auto-drop temperature for GPT-5 2025-08-08 09:33:22 +02:00
Marko Kraemer 010f0ab426
Revert "Revert random PR merged" 2025-07-28 08:14:58 +02:00
Marko Kraemer 454f0d3878
Revert "add support for Google Gemini API integration" 2025-07-28 08:07:42 +02:00
mykonos-ibiza ad9c1dabc8 add support for Google Gemini API integration 
- Introduced GEMINI_API_KEY in setup and environment files.
- Updated SetupWizard to include Google Gemini as a supported LLM provider.
- Adjusted model selection logic to accommodate Gemini models.
- Modified billing and configuration files to handle Gemini model pricing and aliases.
- Updated documentation to reflect the addition of GEMINI_API_KEY.
 2025-07-27 12:49:56 +05:30
LE Quoc Dat 93724249e7 AI: Add to all the related doc : morphllm as a provider; make it optional step when setting up by getting key at this : 
https://morphllm.com/api-keys

when setting up, user should be able to choose to go to morphllm to get a key; or provide open router apikey (if not already in the previous step)
 2025-07-24 13:14:52 +02:00
sharath 52bfed51f6
fix(llm): reorder model provider list for consistency 2025-07-22 05:00:58 +00:00
Sharath 878659aadf
fix(llm): switch kimi k2 provider 2025-07-19 19:50:38 +05:30
sharath 37a3751bbf
enhance(message_tool): update ask function description to include task completion instruction and expand usage scenarios 2025-07-18 16:01:58 +00:00
sharath 13c31d42f6
feat(kimi-k2): add support for Kimi-K2 model with updated token limits and parameters 2025-07-15 13:12:30 +00:00
sharath df871a6eb9
chore(backend): add grok fallbacks 
- Moved fallback model assignment to a dedicated function in `prepare_params`.
- Removed redundant fallback handling from `make_llm_api_call`.
- Updated model constants to include aliases for `xai/grok-4`.
 2025-07-11 15:11:02 +00:00
sharath d3480100f6
Merge branch 'PRODUCTION' 2025-07-11 13:13:34 +00:00
Sharath 8cbaf3d44a
fix(backend): anthropic overloaded fallback failure (#929) 2025-07-11 18:36:15 +05:30
marko-kraemer 957a2c9cbf model fallback, add xai grok 4 2025-07-10 12:58:10 +02:00
Saumya 26a9de5a5d chore: code cleanup and credential profiles 2025-07-09 15:00:30 +05:30
sharath 3fe6a6de61
refactor(llm): streamline cache control application for message content 2025-06-27 15:24:02 +00:00
sharath e98ab052f9
chore(api, billing, llm): update worker count, refine billing tier names, and adjust logging in LLM API calls 2025-06-27 15:12:58 +00:00
sharath fe637b0c51
feat(llm): add fallback parameters for Claude model in LLM service 2025-06-22 19:01:00 +00:00
sharath 1cbc2a63c9
refactor(llm): limit processing of user and assistant messages to a maximum of 4 2025-06-07 10:14:54 +00:00
marko-kraemer c6d13b5187 run, thread_run, response processor, raise errs properly 2025-05-10 03:58:57 +02:00
Marco Vinciguerra e1974c8125 feat: removed unused imports 2025-05-07 15:24:00 +02:00
Korzhavin Ivan 68fafdd385 Refactor agent and tool configurations to support new model names and parameters 
- Adjusted max_tokens logic based on model type, including new conditions for "gpt-4".
- Enhanced MessageTool to include detailed documentation and examples for user interactions.
- Improved ThreadManager to handle new tool choices and reasoning efforts.
- Updated LLM service to incorporate enable_thinking and reasoning_effort parameters.
- Modified chat input component to dynamically manage model selection and thinking options.
 2025-05-03 14:17:46 +02:00
marko-kraemer 7a57ee210a add conf, with local/dev/prod mode 2025-04-24 01:45:58 +01:00
LE Quoc Dat c1514d56a1 improve cache-control 2025-04-23 17:06:40 +01:00
LE Quoc Dat c84ee59dc6 thinking & reasoning 2025-04-18 05:49:41 +01:00
LE Quoc Dat 70882f9292 claude prompt-caching 2025-04-18 04:49:27 +01:00
marko-kraemer 647e10a74b wip 2025-04-17 23:17:48 +01:00
Adam Cohen Hillel f0c3c52cf4 exra 2025-04-15 19:11:09 +01:00
marko-kraemer 57d08c76af wip 2025-04-10 17:02:21 +01:00
marko-kraemer 6eb516e61d native tool call wip 2025-04-07 18:35:40 +01:00
marko-kraemer 3c0634f6bc fix duplicate agent run 2025-04-06 17:45:02 +01:00
Marko Kraemer 7ee68b4282
Refactor to messages table 
* wip

* bedrock support

* wip

* frontend messages table imp

* messages imp
 2025-04-06 11:10:18 +02:00
Marko Kraemer 9edad0073b
v1 full refactor LLM Response Processor (#24) 2025-04-04 17:06:49 +02:00
marko-kraemer 57153f19c8 fix imps 2025-04-01 11:49:35 -07:00
marko-kraemer 486f11ba77 logger 2025-03-31 19:36:26 -07:00
marko-kraemer cbf7ca141e logger v1 2025-03-31 18:26:52 -07:00
marko-kraemer 397c9a7dc6 fe reference wip 2025-03-29 23:48:57 -07:00