Back to timeline

Fri, October 1623:15ToolsModel releasesAPI & pricingOpen sourceModel releases guide

OmniRoute v3.8.52 released: headless mode, quota-share scheduling, and Radar free catalog

Decision Brief

What changedOmniRoute releases v3.8.52, adding a headless service mode, quota-aware scheduling, and an opt-in Radar free catalog overlay.
Why it mattersDevelopers aggregating free quotas with OmniRoute can now run a headless service and share quotas across clients, while the Radar catalog provides fresher free model lists between releases.
Who should careAI coding tool users
Affected stackNo specific stack identified
Source confidenceHigh · Official release / blog / repo

OmniRoute iterated from v3.8.50 to v3.8.52 and beyond, adding three gateway-side capabilities: a headless mode offering serve --headless, quota-aware scheduling with Quota-Share, and a Radar free catalog changed to opt-in, plus release channels as an LTS infrastructure track. Coverage expanded in tandem: providers rose from 352 to 372, standalone chat model IDs from 1320 to 1443, and modality bridging now supports vision, audio, and video. For developers running their own gateway to aggregate multiple free quotas, headless mode lets OmniRoute run as a background service without a desktop UI, while Quota-Share scheduling lets multiple clients or team members reuse the same keys, cutting manual quota allocation. Teams running local open-source models alongside third-party free APIs can use the Radar catalog to roll in new free providers between official releases. Officials stress the headline free-quota figure of roughly 1.62B tokens/month comes from a deduplicated shared-pool catalog count; temporary signup bonuses can push the first month to about 2.22B, and the Radar overlay caps at about 3B/month with no guarantee. With auto model routing, fresh installs may have no keyless targets available, and providers marked tos: avoid (such as OpenCode Free and Kiro) are excluded from auto routing by default.

Summary basis: full article readCompiled from the source scope noted above; the original remains authoritative.

Sources

  • Skill Radar(GitHub 趋势)

    Trending hands-on MCP servers, agent skills, and AI-coding tools discovered from GitHub search momentum.

  • GitHub:diegosouzapw/OmniRoute

Related intel

留言

登入后即可留言,和其他 builder 交换实测心得。

还没有留言,抢头香。