SyslogSolution/prose-contracts
7
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

gpu-self-heal: refresh to current fleet baseline and topology #21

Merged
jerome merged 3 commits from feat/gpu-self-heal-refresh-20260718 into master 2026-07-18 08:36:37 +00:00
Conversation 0 Commits 3 Files Changed 2
+119 -83
abiba-bot commented 2026-07-18 08:07:35 +00:00
Owner
Copy Link
Copy Source

Summary

Updated gpu-self-heal contract to reflect current live fleet state after the 2026-07-17 model swaps.

  • Model assignments synced to current
  • Stable aliases added from gpu-fleet
  • Benchmarks updated to live values
  • Router references removed
  • Prometheus rule replaced with sidecar probe
  • VRAM thresholds updated
  • Rules 11-12 removed
## Summary Updated gpu-self-heal contract to reflect current live fleet state after the 2026-07-17 model swaps. - Model assignments synced to current - Stable aliases added from gpu-fleet - Benchmarks updated to live values - Router references removed - Prometheus rule replaced with sidecar probe - VRAM thresholds updated - Rules 11-12 removed
abiba-bot added 3 commits 2026-07-18 08:07:35 +00:00
gpu-fleet: RTX 5070 swap to HauhauCS Gemma4-12B QAT Uncensored Balanced 5eb6d3bfbd 
- Model: IQ4_NL (6.3GB, 191 tok/s) → Q4_K_M QAT (6.9GB, 87 tok/s)
- MTP draft: Q8_0 (444MB) → tuned draft (242MB), saves 200MB VRAM
- mmproj: F16 → BF16 (same size, matched to new model)
- Benefits: 0/465 refusals, agent-optimized tuning, QAT quality
- Trade: 54% slower generation (acceptable for gpu-light role)
- Config: --parallel 1, --ctx-size 131072, single-slot full 128K
- VRAM: 10.0/12.2GB (82%), healthy headroom
gpu-fleet: RTX 3090 swap to ThinkingCap-Qwen3.6-27B Q4_K_M 31ec70ae36 
- Model: Qwopus Q4_K_M (16GB, 63 tok/s) → ThinkingCap Q4_K_M (15.7GB, 68 tok/s)
- RL-finetuned: 50% fewer thinking tokens, 0.85 MMLU-Pro (vs 0.83 base)
- Self-spec MTP (n=4) REQUIRED for stability — segfaults without it
- Added vision via mmproj (0.9GB) — new capability for this GPU
- VRAM: 20.9/24.6GB (85%), tighter but stable
- Outputs reasoning_content (hidden from Hermes agent)
gpu-self-heal: refresh to current fleet baseline and topology
PR Pipeline — Authorize → Validate → Review → Merge / auth (pull_request) Successful in 4s
Details
PR Pipeline — Authorize → Validate → Review → Merge / validate (pull_request) Successful in 1s
Details
PR Pipeline — Authorize → Validate → Review → Merge / lint (pull_request) Successful in 4s
Details
PR Pipeline — Authorize → Validate → Review → Merge / ai-review (pull_request) Successful in 5s
Details
PR Pipeline — Authorize → Validate → Review → Merge / gate (pull_request) Successful in 1s
Details
bddbb22f03 
- Synced model assignments to 2026-07-17 swaps (ThinkingCap, HauhauCS QAT, Genesis Hermes V3)
- Added stable role-based aliases from gpu-fleet (gpu-dense, gpu-light, strix-moe)
- Updated benchmark baselines to live values (74.9/169.6/62.9 tok/s)
- Replaced router (port 9000) references with LiteLLM + direct routing
- Replaced Prometheus exporter rule with sidecar health probe
- Updated VRAM thresholds to match operational data (300/300/200 MB/h)
- Added response size limit (1MB) to prevent OOM crashes
- Added Lessons L5 (response size crash) and L6 (stable aliases)
- Removed deprecated Rules 11-12 (router-specific distribution balance)
jerome merged commit 14d27a09b5 into master 2026-07-18 08:36:37 +00:00
Sign in to join this conversation.
Reviewers
No Reviewers
Labels
No items
No labels
Milestone
No items
No Milestone
Projects
Clear projects
No projects
Assignees
Clear assignees
abiba-bot jerome (Jerome Kwame Tabiri) kagentz-bot kwame (Kwame Tabiri) mumuni-bot (Mumuni 🦅 (Syslog Falcon)) tanko-bot theo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: SyslogSolution/prose-contracts#21