v0.5.35

antgroup/echomimicv0.5.35Jul 16, 2026by decolua

AI Summary

This release introduces Grok Imagine video generation and CLI tools, enhances routing for Claude models via Copilot, and adds internationalization support for Thai and Persian. It also includes various fixes for API key management, model capabilities, and token handling.

Key Highlights

  • Grok Imagine video generation added to xAI
  • Claude models now routed through Copilot native endpoints
  • Support for Thai (th) and Persian (fa) translations added
  • Kiro direct session cache reuse improved
  • Bulk-add API keys no longer overwrite existing keys

New Features

  • Grok Imagine video generation generation
  • Grok CLI Build setup tool
  • GitHub Copilot Claude model routing
  • Kiro GPT-5.6 model family support
  • RTK header to bypass token savers
  • Quota visibility settings for providers
  • Thai and Persian translations

Full Release Notes

Features:
- xAI: Grok Imagine video generation (/v1/videos) + CLI
- CLI tools: Grok Build setup — writes [model.9router] to ~/.grok/config.toml
- GitHub Copilot: route Claude models through Copilot native /v1/messages
- Kiro: add GPT-5.6 model family (#2596)
- RTK: X-9Router-Token-Saver header to bypass token savers per request
- Providers: quota visibility settings
- Translator: drop temperature for all Claude models
- i18n: Thai (th) + Persian (fa) translations / README

Fixes:
- Providers: bulk-add API keys no longer overwrite existing keys (gap-fill Key N)
- Anthropic: lowercase anthropic-version header to prevent duplication on /v1/messages
- Alicode-intl: use DashScope compatible-mode endpoint so standard keys work
- Grok CLI: align Grok Build with current subscription protocol (#2590)
- Grok CLI: surface expiresAt so proactive token refresh fires (#2546)
- Kiro: improve direct session cache reuse
- Models: populate capabilities for live-catalog LLM models
- Models: list compatible provider models in /v1/models
- Thinking: send explicit thinking:{type:adaptive} alongside output_config.effort
- Translator: strip client_metadata when converting openai-responses to openai

Improvements:
- Perf: skip inactive background services on startup

Docs:
- README: Persian YouTube tutorial