Merge launches Embedded Routing Stack

// 2h agoPRODUCT LAUNCH

Merge launches Embedded Routing Stack

Merge has released its Embedded Routing Stack, a control plane and gateway designed to route LLM requests across major providers like Anthropic, OpenAI, and Google AI while utilizing the customer's own API keys. This solution addresses enterprise data sovereignty by ensuring that personally identifiable information (PII) never leaves the local environment and all requests remain hosted within the EU, facilitating compliant AI agent development for highly regulated industries.

// ANALYSIS

This release highlights a major shift toward model sovereignty and cost transparency, allowing B2B software companies to offer advanced AI agent capabilities without bearing the astronomical API costs of their customers or violating strict GDPR rules.

–BYOK (Bring Your Own Key) billing shifts model token expenses directly to the enterprise customer rather than the SaaS provider.
–EU-hosted regional routing and local PII redacting solve standard compliance bottlenecks that historically barred regulated sectors from leveraging third-party LLMs.
–A single unified control panel reduces developer overhead by eliminating the need to build custom fallback and load-balancing systems for multiple model APIs.

// TAGS

llm-routingapi-gatewayprivacycompliancegdprbyokmodel-sovereigntyagent

DISCOVERED

2h ago

2026-07-17

PUBLISHED

2h ago

2026-07-17

RELEVANCE

8/ 10

AUTHOR

EXM7777

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

NEWS29m ago

Kimi K3 succeeds where Claude Code struggles

Developer levelsio reported that Moonshot AI's Kimi K3 model successfully powered through their Windows XP Simulator to-do list, a task that Claude Code failed to complete over a two-week period. The developer blamed Claude Code's aggressive safety guardrails, which repeatedly downgraded their access from Claude 3 Opus to Claude 3.5 Sonnet, causing constant disruption and wasted time.

MODEL1h ago

Moonshot AI unveils 2.8T MoE Kimi K3

Chinese AI startup Moonshot AI has released Kimi K3, a massive 2.8-trillion-parameter Mixture of Experts (MoE) open-weight model featuring a 1-million-token context window. The release represents a major advancement in open-weight models, showcasing frontier-level capabilities and intensifying the compute race between U.S. and Chinese AI labs.

NEWS1h ago

Kimi K3 faces developer backlash over speed

Moonshot AI recently launched Kimi K3, a massive 2.8 trillion parameter open-weight AI model designed to compete with leading proprietary models from OpenAI and Anthropic. While the model features frontier-level intelligence and a 1-million-token context window, it is hampered by slow performance, averaging only 24 tokens per second with nearly 6 seconds of latency before starting, which disrupts iterative developer workflows like vibe coding.