Portkey Alternative — Routing Without Logging Every Prompt
A Portkey alternative for teams that can't store prompt content. Usage metering without content logs, verifiable in source.
Portkey logs every request. TrustedRouter logs none of them — and proves it.
Portkey is built around observability: it captures every request, response, and trace so you can monitor and govern AI usage. That's exactly what you want for some workloads — and exactly what you can't have for sensitive ones.
TrustedRouter takes the opposite stance. Ordinary synchronous and streaming prompt and output bodies are not persisted, the gateway is hardware-attested, and the whole thing is open source. Batch retention is separate, encrypted, and opt-in. You get routing, fallback, and usage metering without a copy of every prompt sitting in a logging backend.
{
"model_id": "anthropic/claude-sonnet-4.6",
"input_tokens": 12,
"output_tokens": 84,
"cost_microdollars": 218,
"provider": "anthropic",
"region": "us-central1"
}
# No prompt body. No response body. No content.
# Verify the schema in the open-source code.Metering without content.
Token counts, cost, latency, model, region — everything you need for billing and capacity, none of the prompt content.
By construction, not config.
The attested binary never writes request or response bodies. It isn't a toggle you can forget to set; it's a property of the code you can verify.
The gateway features you came for.
30+ providers, automatic rollover on failure, BYOK, prepaid credits — the control-plane conveniences, minus the surveillance.
Pick by what you're allowed to store.
If you need to replay and inspect every request, an observability-first gateway is the right tool. If a stored copy of a prompt is a liability — PHI, privileged material, proprietary code — you need the opposite: a path that provably keeps nothing.
Why we built around proof: “Attestation is All You Need”.
Current routes, prices, privacy, and measured performance.
Catalog facts come from the routes currently configured in TrustedRouter. Performance uses the same cached metadata snapshot as the public leaderboard. Prompts and outputs are not part of these measurements.
| Model | Providers | Context | Input | Output | Privacy | Measured route |
|---|---|---|---|---|---|---|
Anthropic: Claude Opus 4.8anthropic/claude-opus-4.8 |
|
1,000,000 | $5.275/1M | $26.375/1M | varies 5 cited scores | 1538 ms TTFT anthropic · 91 tok/s · 100.00% available · n=3 |
OpenAI: GPT-5.5openai/gpt-5.5 |
|
1,050,000 | $5.275/1M | $31.65/1M | ZDR 3 cited scores | Warming up |
Google: Gemini 3.5 Flashgoogle/gemini-3.5-flash |
+1
|
1,048,576 | $1.5825/1M | $9.495/1M | ZDR | 1278 ms TTFT google-vertex · 107 tok/s · 100.00% available · n=7 |
MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code |
+10
|
262,144 | $0.70685/1M to $1.13096/1M | $3.587/1M to $4.90575/1M | ZDR 5 cited scores | 2262 ms TTFT kimi · 61 tok/s · 100.00% available · n=24 |
Z.ai: GLM 5.2z-ai/glm-5.2 |
+24
|
1,048,576 | $0.7174/1M to $2.434307/1M | $1.5825/1M to $7.029062/1M | E2EE 4 cited scores | 1806 ms TTFT baseten · 128 tok/s · 100.00% available · n=142 |
MiniMax: MiniMax M3minimax/minimax-m3 |
+9
|
524,288 | $0.24265/1M to $0.633/1M | $1.0128/1M to $2.532/1M | ZDR 4 cited scores | 987 ms TTFT fireworks · 220 tok/s · 100.00% available · n=183 |
Browse every modelReview provider policiesOpen the full leaderboardSnapshot 2026-09-17T14:16:36.445Z