Observability Console

Pilot workspace: Northbridge Logistics · local-first routing · hybrid cloud fallback

Cumulative savings vs. cloud-only baseline

Same workload priced at the customer's default path (GPT-4o API) vs. actual Axiom routed cost · mock data
Cumulative $ saved Cloud baseline cost / day Axiom actual cost / day

Intent classification mix

Intent Kernel output · all requests this period

Live request stream streaming

One request, many controls, one signed result — the full inference path made visible
TimeRequestIntentRouteTokensCostSavedp95 msSigned ID
Net savings · rolling 30d
$4,812
▲ 18.2% vs prior 30d
Pilot cost · monthly
$1,650
flat license + GPU amortization
ROI · rolling 30d
292%
priced at 34% of measured savings
Risk-adjusted avoidance
$9,400
17 blocked actions × est. incident cost
Alert rule fired once this week: fallback rate spiked to 9.1% Tue 14:00–15:00 (cloud API latency). Auto-mitigated by raising local specialist confidence threshold. Resolved in 22 min.

Weekly savings trend

Auto-generated line item for the Friday Savings Report

Where the savings come from

Attribution from signed ledger fields · this month

Cost per accepted answer — honest benchmark

Axiom routed SmoLLM stack vs. monolithic local 14B vs. cloud-only · correctness-weighted, benchmark harness v3
PathCost / accepted answerAcceptance ratep95 latencySensitive data exposureVerdict
Axiom routed (0.5B router + 1B/3B specialists + cache)$0.001993.4%640 msnone (local)winner
Monolithic local 14B (no routing)$0.006894.1%1,410 msnone (local)3.6× more expensive
Cloud-only (GPT-4o API)$0.011296.0%890 msleaves perimeter5.9× more expensive
+2.6pp acceptance gap vs cloud is closed on 6.2% of traffic via signed fallback — total cost still 5.9× lower.

Route distribution

Inference Router decisions · tokens never spent on the wrong path

Cache & memory reuse

EventToken / KV-DAG signed reuse — compounding savings
Cache hit rate % Fallback rate %

Model & specialist ledger

Which engine handled what — signed manifests per path
EngineRequestsAvg tokensAvg costAcceptanceAXM delegateVRAM
Qwen2.5-Coder-0.5B-SRD4 (router/draft)21,204312$0.000491.2%0.9 GB
Gemma3-1B-SRD4 (general specialist)9,811540$0.001193.8%gemma-general.axm ✓1.6 GB
TinyLlama-1B-SRD4 (support specialist)5,042468$0.001092.5%support-v2.axm ✓1.4 GB
RAG + signed memory (no generation)8,3660$0.000195.9%
Deterministic tools (calc / lookup / policy)3,9170$0.000099.8%
Cloud fallback (escalation only)3,0111,204$0.014096.4%
AXM delegates load lazily — only 2 of 6 skill packs resident in VRAM this period. Cold-load p95: 310 ms.

Governance events

Pre-execution gates — stopped before action, not logged after

Policy packs

Signed, versioned governance blocks · drift detection on

Gate performance — is control eating efficiency?

Every gate pays rent: overhead measured against savings it creates (blueprint §12)
GateLatency overheadRequests gatedBlockedFalse positivesValue createdStatus
Intent Kernel (risk classify)+18 ms48,3400.4%routing savings $2,910pays rent
CANNOT_MUTATE+3 ms6,120909 destructive actions stoppedpays rent
PII redaction+11 ms8,704312 redactions1.1%0 sensitive leaks to cloudpays rent
Tool-trust ACL+2 ms3,91780.2%8 unauthorized tool calls stoppedpays rent

Signed audit ledger

Every decision signed (HMAC-SHA256) · exportable · this is also the ROI receipt book
Decision IDTimestampActorPath takenPolicy refsCost / savedHMACVerify
Regression pass rate
96.4%
▲ 2.1pp after retro loop
CAS probes run (30d)
1,284
adversarial suite v5
Failures → regression items
46
100% converted · 0 dropped

Adversarial replay results

Every failed or borderline case becomes a regression item — immune response loop
Probe classAttemptsCaught pre-executionCaught post-outputLeakedTrend
Prompt injection (direct)40239840▼ failures
Prompt injection (via tool output)18818161 → regressed▼ failures
PII exfiltration attempt24024000stable
Destructive action social-eng.15415220▼ failures
Policy evasion (paraphrase)300284142 → regressedwatch
Leaked cases auto-enter the regression replay set and are re-tested on every policy-pack version bump.
AXIOM INFERENCE OS · mock observability console · all figures are illustrative demo data, not measurements from a live deployment.
MOCK UI — demo data