# Tokenwise

A drop-in edge proxy for OpenAI, Anthropic, Gemini, Grok, OpenRouter and other LLM APIs that logs cost, tokens, latency and errors per call and suggests fixes such as model swaps, caching and prompt trims, with rules for fallbacks, A/B splits and budget caps.

- Category: Model APIs and gateways · by Tokenwise Labs
- Website: https://tokenwisehq.com/
- Live data: no
- Checked on the official pages: 2026-10-24

## API

- Official API: yes, docs https://tokenwisehq.com/docs
- Key needed: yes
- Price: Paid; limits: Public REST API: 1,000 calls/hour on Indie, 10,000/hour on Pro; proxy plan caps 200,000 (Indie) or 2,000,000 (Pro) requests/month, hard stop at 120%
- Base URL: https://proxy.tokenwisehq.com/
- Formats: json

Example request:

```
https://tokenwisehq.com/api/status
```

## Price and code

- Free tier: No free plan: 7-day trial without a card (200,000 requests/month, 30-day retention); paid from $19/month (https://tokenwisehq.com/pricing)

## Data and license

- License: Tokenwise Terms of Service (https://tokenwisehq.com/terms); covers Use of the proxy, dashboard and public API
- Commercial use OK
- Credit: not stated
- Embedding unclear
- Quote (api_official): "Tokenwise is an HTTP proxy. Point your existing SDK at our base URL with your tw_live_* key — no client library to install, no SDK to maintain." (https://tokenwisehq.com/docs)
- Quote (api_key): "Read-only access to your requests, metrics, evals via tw_api_* bearer tokens." (https://tokenwisehq.com/docs)
- Quote (limits): "Yes. tw_api_* keys hit /api/v1/public/* for workspaces, requests, metrics, and evals. 1,000 calls/hour on Indie, 10,000/hour on Pro." (https://tokenwisehq.com/)
- Quote (limits_proxy): "We hard-stop the proxy at 120% of plan limit to avoid surprise bills." (https://tokenwisehq.com/terms)
- Quote (free_tier): "Trial $0 for 7 days 200,000 / mo 30 days 10" (https://tokenwisehq.com/terms)
- Quote (paid_from): "Indie $19 / mo 200,000 / mo 60 days 10 Pro $79 / mo 2,000,000 / mo 180 days 50" (https://tokenwisehq.com/terms)
- Quote (promo): "Capped at 100 redemptions or July 31, 2026, whichever comes first." (https://tokenwisehq.com/)
- Quote (commercial): "Developers and small teams running LLM features in production: apps and SaaS that call the OpenAI, Anthropic, Google, or other model APIs" (https://tokenwisehq.com/)
- Quote (resale): "Resell raw proxy access as a competing observability product." (https://tokenwisehq.com/terms)
- Quote (status_endpoint): "Real-time edge proxy + app health, refreshed every 30 seconds. Programmatic JSON at /api/status." (https://tokenwisehq.com/docs)
- Quote (keys): "The proxy forwards your provider key upstream and drops it from memory. We never persist provider keys." (https://tokenwisehq.com/)

## Rules for an agent

- Keep the OpenAI/Anthropic SDK and set baseURL to https://proxy.tokenwisehq.com/<provider path> (e.g. /openai/v1, /anthropic, /openrouter/api/v1) plus the Tokenwise key header (docs show x-tw-key, the home page X-Tokenwise-Key); read data via /api/v1/public/* with a tw_api_* bearer token.
- Requests stop at 120% of the monthly plan limit; size the plan to traffic
- Do not resell raw proxy access as a competing observability product
- Prompt payload storage is on by default; turn it off per workspace or tag for sensitive data

Human page: https://agentgid.com/resources/tokenwise/
Source: AgentGid (agentgid.com). Facts are quoted from the site's own pages; check the license page before you ship.