FAQ

Hit a specific symptom — 401 storms, UNPRICED rows, a route that won’t fire? See the symptom-first Troubleshooting page.

Will TokenTriage change my responses?

Not in ledger mode — it observes and costs every request and mutates nothing; the proxied bytes are bytes.Equal to going direct, streaming included. Any internal error fails open to a verbatim passthrough. Routing only rewrites requests once you deliberately climb to route mode. See Core concepts.

Do I have to turn on routing?

No. Routing (and caching, and governance) is opt-in and off by default. Many teams run in ledger mode indefinitely, just to see where the bill goes. See the trust ladder.

Are my prompts stored?

Not by default. Each request writes one JSONL row with a prompt_sha256 and counted features — the prompt text itself is not stored unless you set ledger.redact.prompt: none or enable content capture (dashboard.capture.*, off by default).

Does it phone home?

No. The default posture dials nothing external — the pricing table and dashboard are compiled into the binary. Every phone-out (an external embedder, Redis/Postgres, OTLP export) is refused unless its config carries external: "allow". See Security.

What providers are supported?

TokenTriage ships 95 built-in provider descriptors, and 15 wire-dialect drivers carry all of them: openai, anthropic, bedrock, gemini, anthropic_vertex, elevenlabs, deepgram, assemblyai, replicate, soniox and stability. The generated provider matrix names the test behind each gated cell and prints the ungated ones rather than hiding them; ten of the 95 carry no gated cell at all, and the tests that do exist run against mocks, not against the providers’ live endpoints. Adding an OpenAI-compatible upstream to TokenTriage is a config-only change. See the Providers support matrix.

Do async providers and /v1/batches work?

Yes. The providers that answer a submission with a job id rather than a result (replicate, fal_ai, black_forest_labs, runwayml, stability, assemblyai, soniox) are polled to a terminal state by a built-in poller, and the money lands on a second ledger row when the provider reports it. No descriptor declares the batch endpoint kind, and the openai wire driver still submits, polls and settles POST /v1/batches: declaring an endpoint is metadata for validation, virtual-key scoping and governance matching, not what decides whether a path is served. See Async jobs and Reading a settled cost.

How is cost computed, and what if a model isn’t priced?

From an embedded pricing snapshot stamped source@date#sha7, hand-checkable against the token counts. A model with no price is UNPRICED — a null cost counted and excluded, never $0.00. See The honesty model.

Can it actually prove routing saves money?

It tries, and refuses when it can’t. eval compares the router’s realized quality to a zero-router baseline at the router’s own cost. With only served-model labels it can say routing sits on the frontier; only judged shadow duplication can show a beat; with too few labels it refuses rather than fabricate. See the honesty verdict.

Do I need the SDK?

No — a base-URL swap is enough for costing and routing. The optional SDKs add trace/agent tagging, typed receipts, and a ledger pull when you control the HTTP client.

How do I run it in Kubernetes, or air-gapped?

The canonical path is the Helm chart (one StatefulSet, per-pod PVC, zero Secret objects). A fully self-contained air-gap Compose profile ships too. See Deployment.

What’s the performance overhead?

On a loopback micro-benchmark, proxy overhead is ≈0.065 ms median (a house number on one box); the CI budget is < 2 ms p99. Real provider latency dominates in production. See Benchmarks.

What’s the license?

A commercial licence. TokenTriage is a commercial managed SaaS: there is no free tier, no community edition and no open-source release. Evaluation is a time-limited trial, and after that the product is bought — see Pricing for the three planes and what separates them.

A self-hosted option is under development. Contact [email protected] for early access. Plan entitlement is checked at the API layer at runtime, so there is one build and nothing to swap when your plan changes.

The licence does not grant trademarks: whatever you deploy, do not ship it to others as TokenTriage.

The single static binary is CGO-free and cross-compiles to Linux, Windows, and macOS without a C toolchain.