CodingGenesis archive2026-09
Coding on an 8GB 5060, where tuning the GPU/CPU split matters more than tok/sA daily local coder on an RTX 5060 8GB and 32GB of RAM finds responsiveness under memory pressure matters more than benchmark speed.
Capacity range
Exact reported VRAM
Practice
Model
Quantization
Runtime
Tools / workflow (harness)
Record origin
EXPLORE BY CAPACITY
Start with your machine.
Capacity is the reported total for this setup. Multi-GPU and unified-memory layouts can behave differently.
8 GB or lessLaptops, iGPUs and modest cards9–12 GBEntry-level local inference13–16 GBCurrent mid-range cards17–20 GBRX 7900 XT and compact pro cards21–24 GB3090, 4090 and 7900 XTX class25–32 GB5090 and workstation cards33–48 GBWorkstation and multi-GPU setups49–96 GBLarge multi-GPU and datacenter setups97 GB+Unified memory, servers and clusters
5 resultsEvery record keeps its origin visible.
Batch utilityGenesis archive2026-09
Sensors to reports: deterministic automation with a local model as the extra pair of eyesA raw CSV export from a sensor goes into a local model that produces a report, a pattern the owner treats as a deterministic automation overseen by the model.
Hybrid orchestrationGenesis archive2026-09
An overnight local subagent running unfinished work through cronHermes uses a multi-GPU local server as a coding subagent, with cron inventorying unfinished work overnight and a cloud model reviewing the result.
Hybrid orchestrationGenesis archive2026-09
Frontier plans the tickets, Qwen writes the codeA short but clean division of labor: a frontier model decomposes work into tickets, and a local Qwen model implements them on production code, daily.
CodingGenesis archive2026-09
Rotating between three harnesses to find what actually worksAfter trying Claude Code with task-farming and a Qwen-as-orchestrator experiment, this person settled into two parallel workflows — Orca as the local workhorse, and OpenCode or Bionic for solo agent development.