roboomp cc9fba7513 fix(catalog): normalized kimi k3 reasoning efforts ...

- Derived the mandatory low, high, and max ladder for Kimi K3 across OpenAI-compatible routes.
- Mapped generic requested tiers onto K3 wire values and defaulted the model to max.
- Added request-level regression coverage for LiteLLM-compatible models.

Fixes #5983

2026-07-18 11:36:38 +00:00

..

agent

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

ai

fix(catalog): normalized kimi k3 reasoning efforts

2026-07-18 11:36:38 +00:00

catalog

fix(catalog): normalized kimi k3 reasoning efforts

2026-07-18 11:36:38 +00:00

coding-agent

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

collab-web

chore: bump version to 17.0.1

2026-07-16 04:56:03 +02:00

hashline

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

metaharness

chore: reformat

2026-07-15 00:08:50 +02:00

mnemopi

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

natives

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

snapcompact

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

stats

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

swarm-extension

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

tui

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

typescript-edit-benchmark

feat(harbor-manager): unified benchmark normalization and reporting

2026-07-13 04:21:54 +02:00

utils

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

wire

chore: bump version to 17.0.4

2026-07-18 06:41:08 +02:00

tsconfig.workspace.json

refactor: restructured monorepo TypeScript config and build tasks for unified setup

2026-04-08 17:05:20 +02:00