v0.16.6
minitap-ai/mobile-usev0.16.6Oct 20, 2026by tashfeenahmed
Upgrading?
AI Summary
This release stops the health checker from repeatedly sending empty searches to SerpSearch, Exa, and Parallel keys, which had caused SerpSearch to revoke a user's free account. It also adds ZCode agent support, a new /v1/usage API endpoint, and a "last worked" timestamp on each key in the dashboard.
Key Highlights
- SerpSearch, Exa, and Parallel keys are now validated once at add time instead of every ~5 minutes; error-state keys are re-checked at most hourly
- ZCode integration via `npx freellmapi setup-zcode`, registering FreeLLMAPI in ZCode's shared provider config for Desktop and CLI
- New `GET /v1/usage` endpoint reporting month-to-date requests and tokens split by provider
- Dashboard keys now show a "last worked" time alongside last-checked, distinguishing recent checks from actual successful serving
- New `HEALTH_CHECK_INTERVAL_MS` environment variable to control health-check frequency
New Features
- ZCode agent support with automatic provider registry setup
- `GET /v1/usage` API endpoint (unified key only; client-profile keys receive 403)
- "Last worked" timestamp on dashboard keys
- `HEALTH_CHECK_INTERVAL_MS` configuration (default 5 minutes, minimum 60 seconds)
- Fusion panel ordering now reflects actual request routing under the decision strategy
Full Release Notes
SerpSearch, Exa and Parallel keys are no longer re-checked with empty searches every few minutes. This release also adds ZCode support, a `/v1/usage` endpoint and a "last worked" time on each key. ## Providers - **No more empty-search key checks.** SerpSearch, Exa and Parallel (with a real key) have no account endpoint, so their key check is a search with no query. The health checker used to send it every ~5 minutes for every key, and SerpSearch revoked a user's free account over it. These keys are now checked once, when added. A key in an error state is re-checked at most once an hour, and "Check now" / "Check all" still work (#1608, fixes #1599). ## Agents - **ZCode.** `npx freellmapi setup-zcode --url … --api-key …` adds FreeLLMAPI to ZCode's provider registry (`~/.zcode/v2/provider_config.json`, shared by ZCode Desktop and CLI) and keeps your other providers. ZCode requests show up on the Agents page. CLI 0.8.2 (#1607, #1523). ## Dashboard - **"Last worked" on each key.** Hovering a key's last-checked time now also shows when it last served a request or passed a check, so "checked recently" and "actually working" are no longer the same thing (#1606, #1493). - **Analytics fix.** A filter repeated in the URL (`?provider=a&provider=b`) no longer breaks the recent-calls panel with a 500 (#1597). - **Fusion panel order** now matches how requests actually route under the decision strategy (#1604). ## API - **`GET /v1/usage`.** Month-to-date requests and tokens, split by provider, plus when the window resets. Read from the same ledger as the monthly caps. Unified key only; client-profile keys get 403 (#1600). - **`HEALTH_CHECK_INTERVAL_MS`.** Sets how often keys are health-checked (default 5 minutes, minimum 60 seconds) (#1605, #1593). ## Updating Desktop installers below cover Windows, Apple Silicon and Intel macOS, and Linux. Docker: `ghcr.io/tashfeenahmed/freellmapi:v0.16.6` (`latest` also points here). Existing keys and profiles are retained. One additive migration (`api_keys.last_healthy_at`) runs automatically on start. Full changelog: https://github.com/tashfeenahmed/freellmapi/compare/v0.16.5...v0.16.6 --- ⭐ Like the free router? [Go Premium](https://freellmapi.co/?utm_source=github&utm_medium=release&utm_campaign=premium&utm_content=v0.16.6#pricing) — the live signed catalog, $19/yr, cancel anytime.
Related Repositories
public-apis/public-apis★ 491,314
A collective list of free APIs
NousResearch/hermes-agent★ 256,162
The agent that grows with you
yt-dlp/yt-dlp★ 200,364
A feature-rich command-line audio/video downloader
microsoft/markitdown★ 191,619
Python tool for converting files and office documents to Markdown.
Significant-Gravitas/AutoGPT★ 187,709
AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.