I gave GLM-4.5-Air (106B, open weights) 12 coding tasks through opencode on my RTX 3090. It scored 0%...
GLM-4.5-Air: 0% with opencode, 93% with LangGraph native tool-calling on RTX 3090. For on-premise agents, orchestrator design and tool-tuned models (Qwen3-Coder 30B best) outweigh raw throughput; optimize for watt-cost per correct task.
I gave GLM-4.5-Air (106B, open weights) 12 coding tasks through opencode on my RTX 3090. It scored 0% — never edited a single file.
Same model, same GPU, same tasks, but driven by a ~150-line LangGraph agent instead: 93%.
The model was never the problem. The orchestrator was. Here's the benchmark — including the part nobody else measures, the electricity cost per correct task.
Setup
RTX 3090 (24 GB) + 128 GB RAM, models via ollama, Q4 quants, temp 0.2
See my GLM-5.2 open agent benchmark results. It boosted multi-step tool-use reliability by 22% over Mixtral 8x7B in Node.js,…
Real-world inference benchmarks for coding agents: 31% more TPS than TensorRT-LLM, 2× better TTFT at saturation, and 76% lower…
"It runs on my own GPU, so it's basically free." I believed that until I put a meter on it. So I ran...
I run a homelab with four RTX 3090s — 96 GB of VRAM, 44 CPU cores. For two weeks I tried to make it...
Running AI models locally for code generation used to mean accepting mediocre output. That changed....
The Problem With Choosing a Local Model Everyone has an opinion on which local LLM is...
