Commit graph

545 commits

Author SHA1 Message Date
Alishahryar1
835d0454e8 Fixes for issue 113 and 116 2026-04-18 16:32:31 -07:00
Muhammad Hamid Raza
7468f53ab7
Fix README installation section for uv (#107)
Some checks failed
CI / checks (push) Has been cancelled
2026-03-30 11:08:07 -07:00
Alishahryar1
ec904c6e0c lint
Some checks failed
CI / checks (push) Has been cancelled
2026-03-27 21:49:04 -07:00
Alishahryar1
6dd07d9b6b fix: update test_build_request_body to use enable_thinking=True 2026-03-27 21:48:21 -07:00
Alishahryar1
b75f47b62d Gate NIM thinking params behind NIM_ENABLE_THINKING env var
Mistral models reject chat_template_kwargs, causing 400 errors. Make
thinking params (chat_template_kwargs, reasoning_budget) opt-in via
NIM_ENABLE_THINKING env var (default false) so only models that need it
(kimi, nemotron) receive them.
2026-03-27 21:44:36 -07:00
Alishahryar1
ab0d6aca14 added parallel runs for pytest
Some checks are pending
CI / checks (push) Waiting to run
2026-03-27 21:28:45 -07:00
th-ch
f703a0e403
Implement optional authentication (Anthropic style) (#80)
Some checks are pending
CI / checks (push) Waiting to run
2026-03-27 11:11:47 -07:00
Avishek Behera
587931d279
(doc): Update README with PowerShell and proxy server instructions (#101) 2026-03-27 11:08:43 -07:00
Alishahryar1
812fc5ea3b updated models list 2026-03-27 11:06:23 -07:00
Alishahryar1
2fad4dd4c9 Support both kimi (thinking) and nemotron (enable_thinking) in chat_template_kwargs
Some checks are pending
CI / checks (push) Waiting to run
2026-03-26 12:34:12 -07:00
Alishahryar1
f9e7f65f4c Fix NVIDIA NIM reasoning params for updated API
Replace dropped params (thinking, reasoning_split, include_reasoning,
return_tokens_as_token_ids, reasoning_effort) with the new API format:
chat_template_kwargs.enable_thinking=True and reasoning_budget=max_tokens.
2026-03-26 12:25:04 -07:00
Yuval Dinodia
00038209b2
fix: remove unsupported include_stop_str_in_output NIM param (#95)
Some checks failed
CI / checks (push) Has been cancelled
2026-03-23 11:38:13 -07:00
Ali Khokhar
747262a7ce
Update README.md
Some checks failed
CI / checks (push) Has been cancelled
2026-03-15 12:54:30 -07:00
Alishahryar1
3761780ba1 Updated nvidia nim models list 2026-03-15 12:52:44 -07:00
Alishahryar1
4a59f1f708 Added log line for missed optimization
Some checks failed
CI / checks (push) Has been cancelled
2026-03-11 07:27:16 -07:00
Alishahryar1
55945df1d2 removed logging utils 2026-03-11 07:24:50 -07:00
Xi Gou
4ead059760
update vscode config item name (#81)
Identifier
anthropic.claude-code
Version
2.1.72
Last Updated
2 hours ago
Size
281.35MB
2026-03-11 06:32:20 -07:00
Ali Khokhar
2324be4989
Update README.md
Some checks failed
CI / checks (push) Has been cancelled
2026-03-08 14:35:37 -07:00
Alishahryar1
d679ca2411 Updated Claude.md to point to AGENTS.md 2026-03-08 12:19:18 -07:00
Alishahryar1
5a36a32836 feat: add llama.cpp provider for local anthropic messages API 2026-03-08 10:38:25 -07:00
Alishahryar1
1aedf4763c fix(providers): map httpx exceptions natively and remove type ignores 2026-03-08 08:33:34 -07:00
Alishahryar1
87d8ce1196 feat(lmstudio): route natively to Anthropic /v1/messages endpoint
- Rewrites LMStudioProvider to inherit from BaseProvider
- Passes requests natively to /v1/messages using httpx instead of AsyncOpenAI
- Auto-translates internal ThinkingConfig to Anthropic schema
- Updates .env.example with model routing instructions
- Adjusts test suite for new native integration
2026-03-08 08:17:05 -07:00
Alishahryar1
1959b9a218 perf: use O(1) reverse mapping in CLISessionManager.remove_session 2026-03-08 07:17:51 -07:00
Ali Khokhar
f57598fee3
Move nim_settings from shared base class to NvidiaNimProvider (#78)
Some checks failed
CI / checks (push) Has been cancelled
2026-03-07 22:34:45 -08:00
Ali Khokhar
884ddd77af
Add tests for fcc-init entrypoint (cli/entrypoints.py) (#77)
Some checks are pending
CI / checks (push) Waiting to run
2026-03-07 08:27:11 -08:00
Alishahryar1
fc58b43c5e Update README
Some checks are pending
CI / checks (push) Waiting to run
2026-03-06 22:19:54 -08:00
Alishahryar1
c5e7d6cbb4 Updated uv.lock 2026-03-06 22:13:14 -08:00
Ali Khokhar
c5341ecbbe
Add option for an installable package (#75) 2026-03-06 22:06:33 -08:00
Alishahryar1
ebc8ac461f Updated nvidia models json file
Some checks are pending
CI / checks (push) Waiting to run
2026-03-05 09:26:30 -08:00
Ali Khokhar
a599319dd6
Update README.md
Some checks failed
CI / checks (push) Has been cancelled
2026-03-05 00:19:20 -08:00
Alishahryar1
594dbcb2e3 Set python version to 3.14.0 2026-03-02 05:13:04 -08:00
Ali Khokhar
29125296c7
Update CLAUDE.md 2026-03-01 23:45:43 -08:00
Ali Khokhar
d1652792d3
Update AGENTS.md 2026-03-01 23:45:32 -08:00
Ali Khokhar
4ba2a4cfd3
Update CLAUDE.md 2026-03-01 23:45:02 -08:00
Ali Khokhar
160370268a
Update README with note on new features
Added a note about new features in the README.
2026-03-01 22:30:04 -08:00
Ali Khokhar
63d7f2afe8
Update README 2026-03-01 22:25:06 -08:00
Alishahryar1
ff14baa2d5 Updated README 2026-03-01 22:08:51 -08:00
Alishahryar1
aaa62a2bd7 Relaxed python version requirements 2026-03-01 22:00:34 -08:00
Alishahryar1
2e8b22fa9d Remvoed root insert hack from conftest 2026-03-01 21:57:25 -08:00
Alishahryar1
c1d1368940 Updated README 2026-03-01 21:54:59 -08:00
Alishahryar1
a7d88d5cbd Updated README with per-model mapping, fixed test .env isolation 2026-03-01 21:52:35 -08:00
Alishahryar1
598e21387e Updated README 2026-03-01 21:37:34 -08:00
Alishahryar1
49075b7fa5 Fixed default models 2026-03-01 21:34:01 -08:00
Alishahryar1
ac499cf585 Increased read timeout 2026-03-01 21:33:32 -08:00
Alishahryar1
feba0d456a Updated .env.example 2026-03-01 21:33:17 -08:00
Ali Khokhar
0b324e0421
Per claude model mapping (#66) 2026-03-01 21:32:23 -08:00
Alishahryar1
763c8b62b7 Updated README 2026-03-01 12:47:20 -08:00
Alishahryar1
efb8605258 Updated README 2026-03-01 12:44:40 -08:00
Ali Khokhar
fae8a2a044
Remove over-engineering: drop tree_queue setter, _set_connected(), fi… (#63)
…x cancel_all() TOCTOU

- Remove tree_queue property setter (backward-compat hack; all callers
already migrated to replace_tree_queue()); keep property getter only
- Update 2 remaining tests that still used direct assignment to use
replace_tree_queue()
- Remove _set_connected() 1-line wrapper on DiscordPlatform; assign
_connected directly
- Fix cancel_all() TOCTOU: hold self._lock for the full loop so newly
created trees cannot slip through between the snapshot and cancellation

---------

Co-authored-by: Claude <noreply@anthropic.com>
2026-03-01 12:34:00 -08:00
Ali Khokhar
25b329a3fc
Update README
Removed duplicate VSCode Extension Setup instructions from README.md.
2026-03-01 05:30:30 -08:00