Parth Sareen
91c8e5e1a8
launch: enriched model inventory ( #16230 )
2026-05-21 11:57:20 -07:00
Daniel Hiltgen
4b2d529966
Reduce startup model hydration ( #16215 )
...
* Reduce startup model hydration
Add a lightweight model list cache for tags and launch inventory, while keeping show cache population lazy. This avoids loading every local model at startup on large model stores.
* harden flaky scheduler unit test
* remove extra launch model metadata text
* review comments
* review comments
2026-05-19 15:53:08 -07:00
Bruce MacDonald
e6b1d751f2
codex: omit patch tool type ( #16231 )
...
Including this value can cause schema compatibility issues. It was removed from codex in a new version.
2026-05-19 13:37:05 -07:00
Eva H
56b319f457
launch: add codex model metadata catalog ( #15795 )
...
Co-authored-by: Bruce MacDonald <brucewmacdonald@gmail.com >
2026-05-18 15:26:43 -07:00
Parth Sareen
c28ddc0a7b
launch: codex app restarts ( #16155 )
2026-05-14 12:08:13 -07:00
Parth Sareen
3ad2fa3fb5
launch: update codex app UI copy ( #16157 )
2026-05-14 12:08:08 -07:00
Parth Sareen
ac7295ccab
launch: codex app integration ( #16120 )
2026-05-13 17:11:52 -07:00
Eva H
3af1a008e2
launch/opencode: add image modalities for vision models ( #15922 )
2026-05-12 15:51:46 -04:00
Parth Sareen
f866e7608f
launch: disable Claude Desktop launch ( #16028 )
2026-05-07 10:46:18 -07:00
Parth Sareen
bab59072fb
launch: add plan-aware model gating ( #16027 )
2026-05-06 14:34:26 -07:00
Eva H
7c2c36bda2
cmd/launch: improve integration backup UX ( #15907 )
2026-05-06 11:32:54 -04:00
Parth Sareen
9ba5a04914
launch: claude app ( #15937 )
2026-05-02 19:19:57 -07:00
Pratham Agarwal
8f39fff70b
fix: resolve OpenClaw gateway launch timeout on Windows by enforcing IPv4 loopback ( #15726 )
2026-04-30 22:20:08 -04:00
Parth Sareen
b6447caebc
launch: use vram bytes for model recommendations ( #15885 )
2026-04-29 18:40:14 -07:00
Eva H
bad32c7244
launch/docs: fix title for pool ( #15883 )
2026-04-29 17:18:44 -04:00
Parth Sareen
321cc8a2ba
server/launch: add model recommendations cache endpoint ( #15868 )
2026-04-28 17:09:04 -07:00
Daniel Hiltgen
87288ced4f
New models ( #15861 )
...
* mlx: add laguna model support
* convert: support fp8 safetensors import
Decode HF F8_E4M3 safetensors with block scale companions into GGUF-supported tensor types, and record which output tensors came from FP8 source weights.
Use that source-precision metadata during create quantization: default FP8-sourced GGUFs to Q8_0, keep non-FP8 tensors at their original precision for Q8_0, and promote non-FP8 quantizable tensors to Q8_0 for Q4_K requests.
* ggml: add laguna model support
* server: preserve generate logprobs with builtin parsers
Generate requests were dropping logprob-only chunks whenever a builtin parser buffered visible content. Chat already handled this case, but generate only forwarded chunks with visible response, thinking, or tool-call output.
Keep generate chunks that carry logprobs even when the builtin parser has not flushed visible content yet, and add a regression test that exercises the behavior with a generic thinking parser.
* review comments - perf improvements
* ggml: implement nemotron 3 nano omni
* add poolside integration
* update poolside doc
* adapt to new cache setup
* fix test
* fix test
---------
Co-authored-by: Eva Ho <hoyyeva@gmail.com >
2026-04-28 11:50:12 -07:00
Parth Sareen
590109c835
launch: harden OpenClaw onboarding flow ( #15777 )
2026-04-23 16:47:20 -07:00
Eva H
b4442c6d17
launch: resave managed integration config when live config drifts ( #15776 )
2026-04-23 19:32:36 -04:00
Eva H
85ff8e4a21
launch: keep launch recommended models in a fixed canonical order ( #15750 )
2026-04-23 16:33:00 -04:00
Parth Sareen
160660e572
launch: use bundled OpenClaw ollama web search ( #15757 )
2026-04-22 16:34:19 -07:00
Parth Sareen
21883571b7
launch: replace kimi-k2.5 with k2.6 as top recommended model ( #15737 )
2026-04-21 15:13:20 -07:00
Parth Sareen
8e05d734b9
launch: add kimi cli integration with installer flow ( #15723 )
2026-04-20 15:33:32 -07:00
Parth Sareen
57653b8e42
cmd/launch: show WSL guidance on Windows instead of handing off ( #15637 )
2026-04-16 17:18:04 -07:00
Parth Sareen
a50ce61c54
launch: skip unchanged managed-single rewrite ( #15633 )
2026-04-16 16:20:42 -07:00
Mike Wallio
7d271e6dc9
cmd/launch: add Copilot CLI integration ( #15583 )
...
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com >
Co-authored-by: ParthSareen <parth.sareen@ollama.com >
2026-04-15 17:22:53 -07:00
Eva H
cdddea0592
launch: always list cloud recommendations first ( #15593 )
2026-04-15 13:17:35 -07:00
Parth Sareen
43f90def04
launch: add hermes ( #15569 )
2026-04-15 12:00:23 -07:00
Eva H
698e04a14b
launch: OpenCode inline config ( #15586 )
2026-04-14 15:08:42 -07:00
Eva H
1d9537bc33
launch/openclaw: fix --yes flag behaviour to skip channels configuration ( #15589 )
2026-04-14 13:57:35 -07:00
Eva H
120424d832
Revert "launch/opencode: use inline config ( #15462 )" ( #15568 )
2026-04-13 18:40:17 -07:00
Eva H
5818001610
launch: skip unchanged integration rewrite configration ( #15491 )
2026-04-13 17:18:56 -07:00
Eva H
f3cf6b75fb
launch/opencode: use inline config ( #15462 )
2026-04-13 13:41:31 -07:00
Parth Sareen
80d3744c5d
launch: update openclaw channel message ( #15463 )
2026-04-09 15:20:30 -07:00
Eva H
2a94f03823
launch: add re-run hint to dependency error message ( #15439 )
2026-04-09 09:51:34 -07:00
Eva H
d17f482d50
launch/opencode: detect curl installed opencode at ~/.opencode/bin ( #15197 )
2026-04-08 13:54:51 -07:00
Parth Sareen
4e16f562c0
launch: add openclaw channels setup ( #15407 )
2026-04-08 13:25:27 -07:00
Parth Sareen
55308f1421
launch: update ctx length for glm-5.1 and gemma4 ( #15411 )
...
Also adds glm-5.1 in recommended models
2026-04-08 12:11:50 -07:00
Eva H
d64812eb5d
cmd: improve multi-select sorting and selection status ( #15200 )
2026-04-08 10:39:18 -07:00
Parth Sareen
82f0139587
launch/openclaw: patch approvedScopes baseline for TUI pairing ( #15375 )
2026-04-06 18:00:12 -07:00
Bruce MacDonald
26a58b294c
app: update featured models ( #15373 )
...
Featured models in the app are out of date. Update them to a more recent list of models.
2026-04-06 16:35:35 -07:00
Eva H
a8292dd85f
launch: replace deprecated OPENAI_BASE_URL with config.toml profile for codex ( #15041 )
2026-04-01 11:43:23 -04:00
Jeffrey Morgan
31f968fe1f
cmd: set OpenCode default model in config ( #15127 )
2026-03-29 12:11:36 -07:00
Parth Sareen
8e54823fd3
revert context length warnings change ( #15121 )
2026-03-28 16:43:59 -07:00
Parth Sareen
7c8da5679e
launch: improve multi-select for already added models ( #15113 )
2026-03-28 13:44:40 -07:00
Parth Sareen
6214103e66
launch: auto-install pi and manage web-search lifecycle ( #15118 )
2026-03-28 13:06:20 -07:00
Parth Sareen
b00bd1dfd4
launch: skip context length warning for MLX models and show model name ( #15102 )
2026-03-27 15:01:33 -07:00
Eva H
366625a831
launch: warn when server context length is below 64k for local models ( #15044 )
...
A stop-gap for now to guide users better. We'll add more in-depth recommendations per integration as well.
---------
Co-authored-by: Parth Sareen <parth.sareen@ollama.com >
2026-03-27 00:15:53 -07:00
Eva H
1adfc27f04
launch/vscode: prefer known vs code paths over code on PATH ( #15073 )
2026-03-26 18:06:28 -04:00
Parth Sareen
4a2b9f9dbc
launch: hide cline integration ( #15080 )
2026-03-26 14:33:43 -07:00
