ahmadawais

Introducing GLM-5.3: Built to Code. Ready for Cyber Defense.

- Top-tier coding and agentic capabilities, achieved through post-training on the 743B base model - A major leap in cybersecurity, setting a new standard among open models

Tech Blog: https://z.ai/blog/glm-5.3

We’re launching DeepSeek-V4-Pro today!

Major Agent upgrades with strong production gains! Flexible reasoning effort for V4-Pro & V4-Flash: low for simple tasks, high for daily Agent workflows, max for complex tasks. Native OpenAI Responses API support, optimized for Codex with one-click setup.

V4 Pro is now available on app/web. Try it via “Expert Mode”. V4 Pro is also available via API. Model names remain unchanged—please refer to the API docs for setup details.

i thought that's what we're supposed to do. it has the link too btw.

can't wait for it. i think they'll do 35B variant as well.

> Next week, the open weights of Qwen3.8-Max will be released, and Qwen3.8-27B is also going open-weights to meet you all!

see that's what i was saying.

in v1 we'll have a super strong extension api — week out probably.

Thanks! I've seen many developers use the entire tweet as a prompt to improve their harness. If you trace errors, you can find tool call failures per billion tokens. I've seen like many versions of this on pi, oc, or whatnot. This is already live in our hardness command code for almost two months now and we are now fixing one million tool repairs per one trillion tokens.

also decided to open source Command Code by v1 release, so hopefully end of the month. https://www.youtube.com/watch?v=laEzOCgtK6c (live stream on this)

probably yep, i doubt they'll be able to maintain that well. we did a lot of work improving MiMo's cache thrashing, it was pretty bad early on. Now we're seeing 97% cache rate on their models after lots of prefix work.

hey HN, sharing harness engineering deep dive on tool calling repairs for open models. i've been thinking about why "open model bad at tool calling" is almost always a harness problem, not a model problem. spent time looking at billions of tokens from DeepSeek (and other open models) in our coding agent. ended up building a tool-input repair layer on top of Zod. by the end, DeepSeek V4 Pro was beating Opus 4.7 in 6/10 of our internal evals. the main things that helped:

- most failures came from a small set of recurring schema mistakes - switched from preprocess-then-validate to validate-then-repair - handled some weird cases like markdown auto-links leaking into file paths

full writeup: https://x.com/MrAhmadAwais/status/2050956678502420612

video version (more detailed): https://www.youtube.com/watch?v=f61DCDwvFis

Have you seen any CS bots those are agents. We built an email agent example among many other at https://Langbase.com/docs

This was fun to read. Thanks for sharing. Our memory agents are built somewhat like this but way too much going on in there to make them prod ready.

Couldn’t agree more. Btw what’s your use case?

Spending 100+ hours on this, getting to know people who read this was the point after all. Pretty standard practice everywhere? What did I miss?

I was surprised by the Mistral usage — i think Llama has taken over the open-source LLM scene lately.