GLM-5.3 from Chinese startup Z.ai has tied for the top ranking among open-source AI models with a score of 60 points, matching Kimi K3.
It particularly excels at agentic tasks, with an Elo score of 1,770 on the GDPval-AA v2 benchmark—second only to Claude Opus 5.
At $0.68 per task, it undercuts Kimi K3 by 19 percent, though its open-weights release is delayed by about two weeks while the company implements security controls due to the model's effectiveness at detecting vulnerabilities.
What happened
Z.ai's GLM-5.3 scores 60 points on the Artificial Analysis Intelligence Index, tying with Kimi K3 for the top spot among open models. On the GDPval-AA v2 benchmark, it reaches an Elo score of 1,770—a jump of 246 points from GLM-5.2's 1,524—placing it second overall, behind only Claude Opus 5 (1,855).
Why it matters
GLM-5.3 costs $0.68 per task, making it 19 percent cheaper than Kimi K3 ($0.84), while delivering performance that now matches Western frontier models. The model's strength in agentic tasks (those requiring autonomous decision-making) suggests it can handle complex, multi-step workflows—a capability increasingly valued in enterprise AI.
What to watch
Z.ai is delaying the open-weights release by about two weeks. The company says GLM-5.3 is so effective at detecting security vulnerabilities that it is first strengthening controls and restricting full access to select security partners before broader availability.
Ask the AI about this article →
GLM-5.3 represents a significant performance milestone for open-source AI models. Z.ai's achievement in tying for the top ranking on the Artificial Analysis Intelligence Index while simultaneously capturing second place on the GDPval-AA v2 benchmark suggests that open-source models are narrowing the gap with closed, Western-built frontier models. The model's 246-point jump in agentic task performance—the domain where it made its biggest gains—is particularly noteworthy for businesses exploring autonomous AI systems. At the same time, GLM-5.3's 19 percent cost advantage over Kimi K3 ($0.68 vs. $0.84 per task) signals that competitive pressure on pricing is already reshaping the market for high-performance models.
The delayed open-weights release adds an interesting wrinkle. Z.ai's decision to withhold full access temporarily, citing the model's exceptional ability to detect security vulnerabilities, reflects a deliberate trade-off between speed-to-market and responsible disclosure. By restricting early access to security partners, the company appears to be preventing widespread weaponization of its vulnerability-detection capability before defenses can catch up—a pragmatic stance that may become more common as open models grow more capable.
For example, today's edition would include:
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. Q&As are published on this page for other readers too.
AI system scaling has pushed interconnect requirements inside data centers from chips and boards up to racks…
Chinese large-model developer Z.ai says it can now support large-scale inference using roughly 100,000 domesti…
Analyst Ming-Chi Kuo says Nvidia has revived the Rubin CPX AI accelerator with a substantially redesigned arch…
Palantir Technologies stock has posted multi-year gains, including an 11x return over 3 years
Apple has escalated its legal battle against OpenAI, claiming in a new court filing that OpenAI is actively de…
Samsung Electronics has locked up as much as 70% of its memory production capacity under long-term supply agre…
