LLM

Notes, thoughts, and experiments with Large Language Models.

Pages

• Models Reference — API pricing, benchmarks, and context windows for 50+ models across 11 providers (Anthropic, DeepSeek, OpenAI, Google, GLM, MiniMax, Qwen, Xiaomi, Kimi, Moonshot, OpenCode)
  • SWE-Bench Verified vs Price chart (16 models)
  • SWE-Bench Pro chart (7 models)
  • Full API pricing tables per provider
  • Notes on quotas, caching, and edge cases
• Coding Plans — Subscription plans compared across 9 providers
  • BytePlus ModelArk, GitHub Copilot, GLM/Z.ai, OpenCode Go
  • Claude Code (Anthropic), GPT Codex (OpenAI), Antigravity (Google IDE)
  • Xiaomi MiMo Token Plan, Kimi/Moonshot AI
  • Personal reviews and community sentiment for each
  • Direct buy/subscribe links to all plans

Quick Links