A LinkedIn post from FriendliAI highlights the availability of the GLM-5.1 large language model from Z.ai through what the company describes as a high-performance serverless endpoint. The post cites third-party trackers Artificial Analysis and OpenRouter as indicating that FriendliAI ranks among the leaders in output speed, latency, and response times, as well as in tool calling and structured outputs.
Claim 55% Off TipRanks
- Unlock hedge fund-level data and powerful investing tools for smarter, sharper decisions
- Discover top-performing stock ideas and upgrade to a portfolio of market leaders with Smart Investor Picks
According to the post, GLM-5.1 is positioned for long-horizon, agentic software engineering tasks and is described as outperforming Anthropic’s Claude Opus 4.6 on SWE-Bench Pro and CyberGym benchmarks at a lower cost. For investors, this positioning may signal FriendliAI’s focus on performance-sensitive AI inference workloads, which could support demand from enterprise and developer customers seeking cost-efficient, high-throughput model serving.
The emphasis on benchmarks and time-to-first-token latency suggests that FriendliAI is competing on technical differentiation within the AI infrastructure and model-serving segment. If these performance claims resonate with developers and translate into higher usage of its serverless endpoints, the company could improve revenue scalability and strengthen its standing against rival AI infrastructure providers.
The post also links to a blog article explaining how to run GLM-5.1 on FriendliAI’s platform, implying an effort to drive adoption through technical education and ease of deployment. This kind of ecosystem-building content may help reduce customer onboarding friction, potentially expanding the platform’s user base and enhancing the company’s strategic relevance in the broader AI tooling and cloud services market.