4f95a55c70
Plan abgestimmt mit k9ert im RamaDama-Topic 2026-06-16.
Detection/Execution-Trennung: scripts/wiki-lint.sh ist mechanisch
(kein LLM), Subagent-Spawn nur für Fixes.
Includes:
- scripts/wiki-lint.sh (v0.1, 360 lines, 8 checks, exit 0/1/2)
- scripts/wiki-lint-test.sh (17 asserts, 3 fixtures, all green)
- scripts/test-fixtures/wiki-lint/{clean,messy,broken}-wiki/
- scripts/subagents/{wiki-lint-runner,frontmatter-migration}.md
- wiki/tasks/maintenance-todo.md (WMT-001..006, WMT-000 done)
- AGENTS.md v1.4 (Maintenance-Section hinzugefügt)
- wiki/log.md (Eintrag)
- scripts/wiki-lint-report.md (erster Lint-Run: 46 Pages, 2 Issues)
Erste Lint-Befunde (siehe Report):
- 12 Wiki-Pages ohne Frontmatter
- 14 Stubs (30% Quote)
Co-Authored-By: Hector <hector-bot@botreasury.local>
1.1 KiB
| created | updated | sources | tags | |
|---|---|---|---|---|
| 2026-01-01 | 2026-01-01 |
|
LLM Fusion
This page documents the concept of model fusion and ensemble techniques. The page is linked from memory-system.md so the orphan check should report zero orphans. The line count is high enough to avoid stub status.
Overview
Model fusion combines multiple language models into a single inference path or a routing layer that selects among candidate models. The technique is useful for combining specialised models (coding, reasoning, multilingual) without retraining a monolithic base model. The approach is described in detail in the OpenRouter Fusion whitepaper and the DRACO benchmark study.
Practical Examples
OpenRouter Fusion, DRACO benchmark, and similar approaches show that small ensembles of well-chosen open models can match or exceed much larger frontier models on specific tasks. The economic argument is strong: you pay only for inference, not for the full frontier model on every request. The technique is particularly effective when the candidate models have complementary strengths rather than overlapping coverage of the same tasks.