Edgee launches flat-rate Claude Code Turbo Models
Edgee Turbo Models integrates state-of-the-art open-source LLMs like GLM 5.1 and Kimi K2.7 Code directly into Claude Code at speeds up to 200 tokens per second. Serving full, unquantized open-weight checkpoints on dedicated infrastructure, the service costs a flat $29/month and sets up in minutes.
Flat-rate pricing for AI agents is a brilliant move to counter the unpredictable costs of agentic workflows, though serving top-tier open-source models at high speed under a flat $29 fee will test Edgee's unit economics if users run them heavily.
* Flat-rate pricing solves a major psychological barrier (the "silent tax" of metered billing) for developers using chat-based terminal agents.
* Up to 200 tokens/second on full open-weight models like Kimi K2.7 Code significantly reduces wait times during complex agent reasoning loops.
* By targeting Claude Code directly, Edgee positions itself as an essential utility layer for the growing developer ecosystem around Anthropic's terminal tool.
* Potential concerns include the sustainability of the $29/month pricing under high-volume agent execution, and the reliance on third-party model providers.
DISCOVERED
2h ago
2026-06-16
PUBLISHED
8h ago
2026-06-16
RELEVANCE
AUTHOR
[REDACTED]