The Venice platform now provides anonymous access to Z AI's latest large language model along with availability on OpenRouter and the company's own first party API. GLM-5.3 achieved a score of 60 on the Artificial Analysis Intelligence Index, tying Kimi K3 for the highest intelligence rating among open weights models. The model maintains the 753 billion total and 40 billion active parameter count of GLM-5.2, relying on optimized post-training rather than architecture changes to boost performance in cybersecurity and coding.
Agentic capabilities rose to an Elo of 1770 on the GDPval-AA v2 evaluation, a 246 point jump that places GLM-5.3 second only to Claude Opus 5. The model is priced at $1.40 per 1 million input tokens and $4.40 per 1 million output tokens, though it is less token efficient than its predecessor, using roughly 18,700 output tokens per task. Z AI expects to release the model weights under an MIT license within the next week.
Key sources
- SOURCE@askvenice“GLM 5.3 from @Zai_org is now available on Venice, anonymously.”x.com
- SOURCE@zai_org“Built for coding, defensive cybersecurity, and long-horizon agentic tasks”x.com
- SOURCE@zixuanli_“beginning a broader review of the model weights as we work toward a responsible open-weight release”x.com
- SOURCE@zixuanli_“deliver significant improvements in reasoning, general chat, and specialized domains such as law and finance”x.com
- SUPPORT@designarena“2nd-highest-ranked open-weight model on real-world design tasks”x.com
- SUPPORT@artificialanlys“19% cheaper per task compared to Kimi K3 ($0.84) and 45% cheaper than GPT-5.6 Sol ($1.23)”x.com