LV Relay Launches Free Gateway Grant

// 1h agoINFRASTRUCTURE

LV Relay Launches Free Gateway Grant

LV Relay launched an OpenAI-compatible AI gateway with a testing grant offering new accounts up to ¥10 in platform balance through registration, forum participation, and community signup. The promotion runs through January 1, 2027. [Source](https://linux.do/t/topic/3044534)

// ANALYSIS

Free quota is a smart wedge in the crowded LLM gateway market, but developers should treat this as an evaluation budget rather than production infrastructure.

–OpenAI compatibility lowers migration friction for existing SDKs, pipelines, and agent workflows.
–The grant combines ¥3 for registration, ¥2 for posting a user ID, and ¥5 for joining the support group.
–The announcement does not publish a detailed model catalog, latency target, uptime commitment, or data-retention policy.
–Test routing quality, rate limits, and reliability before moving real workloads onto the gateway.
–Keep provider-level fallbacks or your own API credentials to avoid lock-in.

// TAGS

lv-relayllminferenceapidevtoolhosted-servicepricing

DISCOVERED

1h ago

2026-10-20

PUBLISHED

1h ago

2026-10-20

RELEVANCE

8/ 10

AUTHOR

dasepmochly

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

FUNDING55m ago

OpenAI Acquires Orchestra Bio for Rosalind

OpenAI has acquired Orchestra Bio, an agentic platform that helps biopharma teams coordinate drug-development programs. Its team and technology will join Rosalind to connect scientific findings with study planning, operational tracking, and next-step decisions.

UPDATE57m ago

Relizy 1.8.0 adds Vercel AI SDK

Relizy 1.8.0 adds an ai-sdk provider for rewriting release notes and social posts with models from the Vercel AI SDK, including OpenAI, Google, Mistral, and AI Gateway models. It also fixes previous-release detection so independent monorepo releases no longer omit commits.

MODEL1h ago

Ornith-1.5-122B Hits 84.5% SWE-bench

Ornith-1.5-122B is a 122B mixture-of-experts model claiming 84.5% on SWE-bench Verified and 78.5% on BrowseComp while running locally on a single DGX. Its MIT-licensed weights and OpenAI-compatible tooling target high-end private coding and agent workloads.