steve/gadfly
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(reusable): add the 4090 Ti (qwen3.6-27b via llama-swap) to the default swarm #13

Merged
steve merged 2 commits from feat/ragnaros-4090 into main 2026-06-28 05:01:50 +00:00
Conversation 8 Commits 2 Files Changed 3
+19 -9

2 Commits

Author SHA1 Message Date
Steve Dudenhoeffer 4cef1db686 fix: use un-hyphenated 'llamaswap' provider for the ragnaros endpoint
Build & push image / build-and-push (pull_request) Successful in 8s
Details
Adversarial Review (Gadfly) / review (pull_request) Successful in 5m31s
Details 
PR #13's self-review showed the pinned image (sha-c342bdb) rejects the
'llama-swaps' provider spelling in GADFLY_ENDPOINT (its endpointProvider only
accepts 'llamaswap'; the hyphenated aliases were added to the binary later).
Switch GADFLY_ENDPOINT_RAGNAROS to llamaswap|https://... so the 4090 Ti
endpoint registers and ragnaros/qwen3.6-27b resolves.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
 2026-06-28 00:37:49 -04:00
Steve Dudenhoeffer 2b02cbb4ba feat(reusable): add the 4090 Ti (qwen3.6-27b via llama-swap) to the default swarm
Build & push image / build-and-push (pull_request) Successful in 4s
Details
Adversarial Review (Gadfly) / review (pull_request) Successful in 4m21s
Details 
Adds a local GPU reviewer to the shared default:
- models += ragnaros/qwen3.6-27b
- GADFLY_ENDPOINT_RAGNAROS=llama-swaps|https://llama-swap.ragnaros.dudenhoeffer.casa
  (plain LAN URL, no credential; registers provider "ragnaros")
- provider_concurrency ragnaros=1, provider_lens_concurrency ragnaros=1
  (one model, one lens at a time — a single local GPU)

Inherited by all @v1 consumers (mort/executus/majordomo) once v1 moves.
Comments + README + CLAUDE.md updated.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
 2026-06-28 00:23:45 -04:00