d14d816b22
Make markdown live vertical_overflow and refresh_interval configurable ( #785 )
...
Co-authored-by: james.z <james.zheng@sjfood.com >
Co-authored-by: Farkhod Sadykov <farkhod.sadykov@gmail.com >
2026-07-02 07:17:03 +02:00
6783862a9a
Make markdown vertical_overflow configurable ( #779 )
2026-06-23 04:37:42 +02:00
17be969232
Update litellm integration ( #762 )
...
- Bump litellm from 1.42.5 to 1.83.4
- Keep api_key in additional_kwargs for litellm completion calls
- Remove outdated Docker + Ollama section from README
- Minor wording improvements in README
- Fix unicode dash in REPL help text
2026-05-06 05:22:19 +02:00
4ea2f834cf
Migrate to OpenAI v2 ( #744 )
...
* Migrate to OpenAI v2 package
* Remove Click direct package dependency
* Minor handler improvements
* Change README.md demo video
* Version bump, release 1.5.0
2026-01-28 01:52:29 +01:00
880e7db0d0
Add check for empty choices in response chunks ( #738 )
2026-01-27 22:11:00 +01:00
9615dfbec8
Minor lint fixes
2025-07-17 04:19:16 +02:00
b7cad0bd85
Retain system role message in cache file
2025-04-11 10:40:09 +02:00
8f93f280ce
Message cache length fix ( #683 )
2025-04-08 12:24:29 +02:00
859c97915a
Preserve system message in the chat message cache ( #669 )
2025-04-08 12:06:38 +02:00
8fbd94f60b
Fix --show-chat and --repl to respect --no-md ( #513 )
2025-02-17 01:17:31 +01:00
aac2f5461b
Option for LiteLLM to manage API keys ( #604 )
2024-08-29 23:27:44 +02:00
b087d73bf1
Minor bug fixes, documentation, release 1.4.4 ( #607 )
2024-08-11 00:02:01 +02:00
bbd78d0190
Adopt new function calling that supports Groq and other models ( #569 )
...
Co-authored-by: Hejia Zhang <hejiazhang2023@outlook.com >
2024-08-10 19:19:57 +02:00
7b619e1fa5
Release version 1.4.1 ( #532 )
2024-04-06 18:09:10 +02:00
be138cbd77
Chat and REPL modes output fix after function calls ( #525 )
2024-04-01 07:35:16 +02:00
bb63eba9c5
Keyboard interrupt while streaming ( #521 )
2024-03-24 20:46:16 +01:00
90f3f4c294
Use OpenAI by default make LiteLLM optional ( #488 )
2024-02-22 03:43:56 +01:00
779749f304
Markdown option and config variable ( #481 )
2024-02-20 23:55:49 +01:00
ecb7b26e15
New config variable API_BASE_URL and minor bug fixes ( #477 )
2024-02-17 02:58:55 +01:00
f4dc37f925
Markdown formatting for chat history ( #444 )
2024-02-11 20:03:50 +01:00
1cb61dee0a
Ollama integration 🦙 ( #463 )
2024-02-09 23:18:39 +01:00
ad6d297b28
Major refactoring and optimisation ( #462 )
...
* Major refactoring and optimisation.
* Removed redundant code.
* Removed caching for --chat conversations.
* Fixing mypy typing.
* Printer class to handle LLM output formatting.
* Spinner and better handling when DISABLE_STREAMING is True.
2024-01-31 03:55:39 +01:00
c48926a4a4
Non-interactive mode for --shell and integration changes ( #455 )
...
* Added --interaction that works with --shell option.
* Changed shell integrations to use new --no-interaction option.
* Moved shell integrations into dedicated file integration.py.
* Changed --install-integration logic to install integrations without downloading sh scripts.
* Removed validation for PROMPT argument, empty string by default.
* Fixing an issue when sgpt is being called from non-interactive shell environments.
* Fixed and optimised Dockerfile.
* README.md improvements, and new feature examples.
* New funny demo video.
2024-01-28 05:03:44 +01:00
361d1eabf8
Improved tests, REPL stdin, documentation ( #452 )
...
* Improved mocked response tests.
* REPL mode can accept stdin as initial prompt.
* Readme and Contributing documentation improvements.
2024-01-19 02:16:46 +01:00
b77b7d7c2f
Improved tests with mocked API responses ( #442 )
...
* Changed name of --top-probability parameter to --top-p.
* Fixed bug in --repl --shell when describing shell command.
2024-01-15 05:24:31 +01:00
20ff0f2eeb
Added OpenAI function calls ( #427 )
2024-01-09 02:52:41 +01:00
4b670cf8af
Added OpenAI SDK dependency ( #414 )
2023-12-23 04:05:34 +01:00
c8e279c512
REPL conversation bug fix and minor improvements ( #410 )
2023-12-22 01:38:29 +01:00
b0d43463af
REPL multiline documentation and improvements ( #401 )
2023-12-19 21:59:01 +01:00
3eac96b5d7
Render markdown for default output ( #400 )
...
Co-authored-by: Jaycen Horton <jaycen@nori.com >
2023-12-19 21:29:21 +01:00
c0fc39dc4f
System role optimisations for GPT4+ models ( #398 )
2023-12-19 16:53:15 +01:00
f50d544086
Multiline input in REPL ( #393 )
2023-12-12 23:14:14 +01:00
d2a43c6723
Add option to disable response streaming ( #290 )
2023-06-21 00:06:27 +02:00
0129e59db6
Option to describe --shell commands when executing ( #249 )
2023-05-21 19:45:51 +02:00
a054a90095
Custom roles ( #183 )
...
* Implementing custom roles.
* Implementing system role messages.
* Small stdin change with new lines at the end.
* Small option name fix --list-chats.
* Refactoring module imports.
* Adding tests for roles.
2023-04-16 18:27:49 +02:00
12134c2c95
Migrating to TOML new linting and stricted typing ( #165 )
...
* pyproject.toml
* sdist and bdist specific files.
* CONTRIBUITNG.md.
* New linters and CI/CD.
* Mypy stricted types.
* Shell scripts to run linters and tests.
2023-04-15 01:22:23 +02:00
8fe43f562d
Option to change completion color ( #157 )
...
* Optimized config module with Config class.
* Added config option to change OpenAI completion color in terminal.
2023-04-10 12:18:21 +02:00
c8da76a2c4
Native $SHELL commands and better Windows support ( #149 )
...
* Native $SHELL command execution instead of default /bin/sh.
* Added dependency pyreadline3 for Windows systems.
* Fixed a bug when sgpt couldn't recognize powershell or cmd on Windows.
* Better integration with powershell and cmd on Windows.
2023-04-07 20:28:31 +02:00
096b690e63
REPL mode for chat sessions ( #94 )
2023-04-03 02:28:34 +02:00
763fef15d3
Refactoring, more tests, lint fixes, optimise
...
* Refactoring, more tests, lint fixes, optimise
* Lint CI fixes
* Changing version, readme help fixes
* Validation rule, validation tests, minor improvements
2023-04-02 02:52:55 +02:00
、Idea