A Reddit user, a heavy Claude Code user with over 12 billion tokens consumed, shared their experience with GLM 5, a model tested via OpenCode with the Zen plan.

Initial Tests and Results

After an initial unconvincing trial with Kimi K2.5, the user compared GLM 5 and Claude Code on two software development tasks. The first, a simple dashboard for tracking an inventory, saw Claude Code slightly ahead. The second, more complex task involved creating a real-time chat application with web sockets. In this scenario, GLM 5 unexpectedly outperformed Claude Code, which initially did not offer message streaming functionality.

Evaluation and Feedback

The user rated GLM 5 as superior based on their criteria and provided detailed feedback to both development teams. Even after implementing the suggested changes, GLM 5 maintained an advantage. This result led the user to question the true extent of GLM 5's capabilities and to invite the community to propose even more complex development tasks to fully assess the differences between the two models.

For those evaluating on-premise deployments, there are trade-offs to consider carefully. AI-RADAR offers analytical frameworks on /llm-onpremise to support these evaluations.