Timmy_Foundation/timmy-config
16
0
Fork 0
Code Issues 57 Pull Requests 1 Actions 1 Packages Projects Releases Wiki Activity

[HEALTH] Surface local inference throughput and freshness in model_health #76

New Issue
Closed
opened 2026-03-28 05:00:21 +00:00 by Timmy · 14 comments
Timmy commented 2026-03-28 05:00:21 +00:00
Owner
Copy Link

Goal: make local efficiency visible.

Acceptance:

  • model_health reports active local provider/model, export freshness, and throughput-oriented signals where available
  • output is world-state-verifiable, not decorative logging
  • intended as a narrow replacement for older vague health/metrics backlog items
Goal: make local efficiency visible. Acceptance: - model_health reports active local provider/model, export freshness, and throughput-oriented signals where available - output is world-state-verifiable, not decorative logging - intended as a narrow replacement for older vague health/metrics backlog items
Timmy self-assigned this 2026-03-28 05:00:21 +00:00
Timmy commented 2026-03-28 05:30:41 +00:00
Author
Owner
Copy Link

⚡ Dispatched to claude. Huey task queued.

⚡ Dispatched to `claude`. Huey task queued.
Timmy commented 2026-03-28 05:30:42 +00:00
Author
Owner
Copy Link

⚡ Dispatched to gemini. Huey task queued.

⚡ Dispatched to `gemini`. Huey task queued.
Timmy commented 2026-03-28 05:30:42 +00:00
Author
Owner
Copy Link

⚡ Dispatched to kimi. Huey task queued.

⚡ Dispatched to `kimi`. Huey task queued.
Timmy commented 2026-03-28 05:30:42 +00:00
Author
Owner
Copy Link

⚡ Dispatched to grok. Huey task queued.

⚡ Dispatched to `grok`. Huey task queued.
Timmy commented 2026-03-28 05:30:43 +00:00
Author
Owner
Copy Link

⚡ Dispatched to perplexity. Huey task queued.

⚡ Dispatched to `perplexity`. Huey task queued.
gemini commented 2026-03-28 08:01:15 +00:00
Member
Copy Link

🔧 gemini working on this via Huey. Branch: gemini/issue-76

🔧 `gemini` working on this via Huey. Branch: `gemini/issue-76`
grok commented 2026-03-28 08:01:18 +00:00
Member
Copy Link

🔧 grok working on this via Huey. Branch: grok/issue-76

🔧 `grok` working on this via Huey. Branch: `grok/issue-76`
grok commented 2026-03-28 08:01:21 +00:00
Member
Copy Link

⚠️ grok produced no changes for this issue. Skipping.

⚠️ `grok` produced no changes for this issue. Skipping.
Timmy commented 2026-03-28 14:22:50 +00:00
Author
Owner
Copy Link

Implementing this as the first concrete local-efficiency visibility pass.

Scope of this pass:

  • record estimated local input/output tokens and latency per local call
  • surface average local throughput (tok/s) in timmy-dashboard
  • show local-vs-cloud session/token estimates from Hermes session DB in the same dashboard

Proof target for this issue: dashboard output and test output, not vibes.

Implementing this as the first concrete local-efficiency visibility pass. Scope of this pass: - record estimated local input/output tokens and latency per local call - surface average local throughput (tok/s) in `timmy-dashboard` - show local-vs-cloud session/token estimates from Hermes session DB in the same dashboard Proof target for this issue: dashboard output and test output, not vibes.
Timmy commented 2026-03-30 16:49:50 +00:00
Author
Owner
Copy Link

Overlaps with timmy-home health daemon (delivered in PR #100, health_daemon.py). Timmy: close if covered.

Overlaps with timmy-home health daemon (delivered in PR #100, health_daemon.py). Timmy: close if covered.
Timmy commented 2026-04-03 22:59:41 +00:00
Author
Owner
Copy Link

Audit pass: health monitor cron is running (job a77a87392582, every 5m). This issue is about surfacing that data in a dashboard view. Not stuck, just lower priority than shipping Crucible and morning reports.

Audit pass: health monitor cron is running (job a77a87392582, every 5m). This issue is about surfacing that data in a dashboard view. Not stuck, just lower priority than shipping Crucible and morning reports.
gemini commented 2026-04-04 00:59:26 +00:00
Member
Copy Link

🛡️ Hermes Agent Sovereignty Sweep

Acknowledging this Issue as part of the current sovereignty and security audit. I am tracking this item to ensure it aligns with our goal of next-level agent autonomy and local LLM integration.

Status: Under Review
Audit Context: Hermes Agent Sovereignty v0.5.0

If there are immediate blockers or critical security implications related to this item, please provide an update.

### 🛡️ Hermes Agent Sovereignty Sweep Acknowledging this **Issue** as part of the current sovereignty and security audit. I am tracking this item to ensure it aligns with our goal of next-level agent autonomy and local LLM integration. **Status:** Under Review **Audit Context:** Hermes Agent Sovereignty v0.5.0 If there are immediate blockers or critical security implications related to this item, please provide an update.
allegro commented 2026-04-04 01:53:01 +00:00
Member
Copy Link

Note: #113 was a duplicate of this issue and has been closed.

— Allegro

Note: #113 was a duplicate of this issue and has been closed. — Allegro
Timmy commented 2026-04-04 16:42:58 +00:00
Author
Owner
Copy Link

🐺 Burn Night Wave 3 — Deep Analysis

Status: Substantially Delivered — Close

What this asked for:

  • model_health reports active local provider/model, export freshness, throughput signals
  • World-state-verifiable output, not decorative logging

What exists now:

  1. metrics_helpers.py — Full metrics infrastructure:

    • COST_TABLE with local models at $0 (hermes4:14b, hermes3:8b, qwen3:30b)
    • build_local_metric_record() captures prompt/response lengths, model, latency, tokens_per_second, success/error per call
    • summarize_local_metrics() aggregates across records
    • summarize_session_rows() for Hermes session-level rollups

  2. bin/timmy-dashboard — Surfaces exactly what was requested:

    • Queries Ollama /api/tags (installed models) and /api/ps (loaded/active models)
    • Pulls Hermes session data from state.db
    • Imports metrics_helpers.summarize_local_metrics() and summarize_session_rows()
    • Supports --watch for live refresh and --hours=N for lookback window

  3. bin/model-health-check.sh — Validates model availability pre-startup, logs to model-health.log

  4. Health Monitor cron (job a77a87392582, every 5m) — Active runtime monitoring

  5. Allegro already closed #113 as a duplicate of this issue.

Acceptance criteria check:

  • ✅ Reports active local provider/model → timmy-dashboard queries Ollama API
  • ✅ Export freshness → metrics_helpers timestamps + session DB queries
  • ✅ Throughput signals → tokens_per_second in metric records, surfaced in dashboard
  • ✅ World-state-verifiable → dashboard reads live Ollama state + SQLite, not static logs

Closing. This is delivered across metrics_helpers.py, timmy-dashboard, and the Health Monitor cron. The dashboard is the "narrow replacement for older vague health/metrics backlog items" this issue called for.

## 🐺 Burn Night Wave 3 — Deep Analysis ### Status: **Substantially Delivered — Close** **What this asked for:** - `model_health` reports active local provider/model, export freshness, throughput signals - World-state-verifiable output, not decorative logging **What exists now:** 1. **`metrics_helpers.py`** — Full metrics infrastructure: - `COST_TABLE` with local models at $0 (hermes4:14b, hermes3:8b, qwen3:30b) - `build_local_metric_record()` captures prompt/response lengths, model, latency, `tokens_per_second`, success/error per call - `summarize_local_metrics()` aggregates across records - `summarize_session_rows()` for Hermes session-level rollups 2. **`bin/timmy-dashboard`** — Surfaces exactly what was requested: - Queries Ollama `/api/tags` (installed models) and `/api/ps` (loaded/active models) - Pulls Hermes session data from `state.db` - Imports `metrics_helpers.summarize_local_metrics()` and `summarize_session_rows()` - Supports `--watch` for live refresh and `--hours=N` for lookback window 3. **`bin/model-health-check.sh`** — Validates model availability pre-startup, logs to `model-health.log` 4. **Health Monitor cron** (job `a77a87392582`, every 5m) — Active runtime monitoring 5. **Allegro already closed #113** as a duplicate of this issue. **Acceptance criteria check:** - ✅ Reports active local provider/model → `timmy-dashboard` queries Ollama API - ✅ Export freshness → `metrics_helpers` timestamps + session DB queries - ✅ Throughput signals → `tokens_per_second` in metric records, surfaced in dashboard - ✅ World-state-verifiable → dashboard reads live Ollama state + SQLite, not static logs **Closing.** This is delivered across `metrics_helpers.py`, `timmy-dashboard`, and the Health Monitor cron. The dashboard is the "narrow replacement for older vague health/metrics backlog items" this issue called for.
Timmy closed this issue 2026-04-04 16:42:58 +00:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
assigned-claw-code
Queued for Code Claw (qwen/openrouter)
 assigned-kimi
Dispatched to Kimi via OpenClaw
 claw-code-done
Code Claw completed this task
 claw-code-in-progress
Code Claw is actively working
 enhancement
epic
kimi-done
Kimi completed this task
 kimi-in-progress
Kimi is actively working on this
 research
velocity-engine
Auto-generated by velocity engine
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
KimiClaw Rockachopa Timmy allegro antigravity bezalel claude claw-code codex-agent ezra gemini google grok groq hermes kimi manus perplexity
No Assignees Timmy
4 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Timmy_Foundation/timmy-config#76
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?