Glm-4.6: Advanced Agentic, Reasoning and Coding Capabilies

Very respectable. If Sonnet 4.5 was delayed by a day, GLM 4.6 would have been the obvious best choice for agentic coding... up there with gpt-codex.

The z.ai monthly plan looks like a steal if you're working on personal hobby projects instead of paying the $200/mo price for Claude Code.

GLM 4.5 is a great budget option for me. After cancelling Claude sub ($20 USD one), I've been trying Gemini 2.5 Pro through Gemini CLI, and it was okay for landing pages and UI. However, I switched to GLM through Claude Code using their cheapest subscription and it's been pretty good so far. It feels more creative than Gemini when it comes to UI, not Claude Sonnet level (can't compare with Opus), but it's 90% there. It also follows instructions better than Gemini, and makes fewer or no syntax errors. Gemini through Gemini CLI would somewhat often completely mess up Astro files, or be a bit lazy. No such problems on AI Studio though.

A few days I got their quarterly promo plan ($9 USD), so I'm excited to try 4.6.

    alias zcc='ANTHROPIC_BASE_URL="https://api.z.ai/api/anthropic" ANTHROPIC_AUTH_TOKEN="your z.ai api key" claude --model GLM-4.6'

GLM 4.5 for me one of the best coding model when running with agents such as Claude Code and Opencode (excluding Claude and GPT-5).

Looking forward to try GLM 4.6.

But Z AI should do like KIMI¹ and launch a verifier to see how GLM models behave on 3rd party providers.

[1] https://github.com/MoonshotAI/K2-Vendor-Verfier

The most used opensource agentic coding model, Qwen 3 Coder, is completely missing from all the comparisons. That's what everyone who uses these models wants to compare it with. A missed opportunity.

You need to update your pricing page, still saying GLM-4.5 only

It would be great if people that are using GLM-4.6 within Claude Code could chime in with their experience in using it. I'm a MAX subscriber and having extra cheap tokens would be very helpful.