gen_6h1bnjq2rr

Decision timeline

1. admission+0 ms

   chat request by prin_admin under profile @company/general-fast (requested "@company/general-fast")

2. PRE_ROUTE+0 ms
   allowsize-budgetno findings [0.00ms]
3. PRE_ROUTE+0 ms
   allowprompt-injection-heuristicno findings [0.00ms]
4. PRE_MODEL+0 ms
   allowpattern-dlpno findings [0.00ms]
5. PRE_MODEL+0 ms
   allowsecrets-scannerno findings [0.00ms]
6. cache+0 ms

   cache bypassed (policy off or non-cacheable request)

7. route+0 ms

   selected openai/gpt-5-mini via openai/acct_openai_pool_a_euw1 on node_cf_euw1_01 (B-tier, eu-west-1) — objective=latency, 12 candidates scored, 11 in fallback chain, 0 hard-filter rejections [router router_1.0, policy rp_default/@company/general-fast@v4]

   12 candidates considered12 survived hard filters0 rejectedobjective: latencyfallback chain: 11router_1.0 · rp_default/@company/general-fast@v4

   Score breakdown — top 5

   #	Candidate	quality	cacheSavings	contractValue	completionProbability	cost	latencyPenalty	shadowPrice	retryRisk	affinityPenalty	guardrailCost	Total
   1	openai/gpt-5-mini via openai · node_cf_euw1_01 selected	+0.344	0.000	0.000	+0.392	-0.014	-0.019	-0.010	-0.005	0.000	-0.040	0.648
   2	openai/gpt-5-mini via openai · node_cf_use1_01	+0.344	0.000	0.000	+0.398	-0.014	-0.017	-0.010	-0.001	0.000	-0.060	0.639
   3	openai/gpt-5-mini via openai · node_cf_use1_01	+0.344	0.000	0.000	+0.398	-0.014	-0.041	-0.010	-0.001	0.000	-0.040	0.636
   4	openai/gpt-5-mini via openai · node_sg_dc_01	+0.344	0.000	0.000	+0.396	-0.014	-0.041	-0.010	-0.003	0.000	-0.040	0.632
   5	openai/gpt-5-mini via openai · node_sg_dc_01	+0.344	0.000	0.000	+0.396	-0.014	-0.041	-0.010	-0.003	0.000	-0.060	0.612

8. attempt #1+0 ms

   #1 openai · account acct_openai_pool_a_euw1 · node node_cf_euw1_01 — succeeded in 1256ms

   succeededopenaiacct_openai_pool_a_euw1pool pool-anode_cf_euw1_01Tier B1.26 s

   14 in / 292 out

9. POST_MODEL+0 ms
   allowtoxicity-liteno findings [0.00ms]
10. PRE_RETURN+0 ms
    allowschema-validatorno findings [0.00ms]
11. settlement+1256 ms

    settled $0.000588 — 14 in / 292 out tokens (306 total)

    Ledger entries

    • reserve$0.0007budget:bud_proj_corereserve est max for @company/general-fast
    • settle$0.0006budget:bud_proj_coresettle actual provider usage (306 tok, internal charge $0.000676)
    • settle−$0.0006account:acct_openai_pool_a_euw1credit provider account acct_openai_pool_a_euw1 (double-entry counterpart)
    • refund−$0.0001budget:bud_proj_corerefund unused reservation

    Cost layers (memo §15.3)

    Provider list

    $0.0006

    Contract-adjusted

    $0.0006

    Internal charge

    $0.0007

    Contract savings

    $0.00

    contract ct_openai_metered (metered)

12. complete+1256 ms

    status=succeeded, model=openai/gpt-5-mini, 1256ms end-to-end

fabric_metadata

The route-explanation envelope an API caller receives with X-OpenRouter-Metadata: enabled or routing.explain, reconstructed from this generation record.

{
  "requested": "@company/general-fast",
  "resolved_model": "openai/gpt-5-mini",
  "strategy": "latency",
  "policy_version": "rp_default/@company/general-fast@v4",
  "router_version": "router_1.0",
  "provider": "openai",
  "region": "eu-west-1",
  "account_pool": "pool-a",
  "node": {
    "ref": "node_cf_euw1_01",
    "trust_tier": "B"
  },
  "cache": {
    "response": "BYPASS",
    "prompt_affinity": "MISS",
    "predicted_cached_tokens": 0
  },
  "guardrails": [
    {
      "stage": "PRE_ROUTE",
      "action": "allow",
      "profile": "gp_baseline",
      "evaluations": 2
    },
    {
      "stage": "PRE_MODEL",
      "action": "allow",
      "profile": "gp_baseline",
      "evaluations": 2
    },
    {
      "stage": "POST_MODEL",
      "action": "allow",
      "profile": "gp_baseline",
      "evaluations": 1
    },
    {
      "stage": "PRE_RETURN",
      "action": "allow",
      "profile": "gp_baseline",
      "evaluations": 1
    }
  ],
  "attempts": [
    {
      "provider": "openai",
      "status": 200
    }
  ],
  "routing_reason": [
    "selected openai/gpt-5-mini via openai (latency objective, score 0.6479)"
  ]
}