DeepSeek-V4-Flash Agent Eval: Completes 3D Task for $0.07
cedric_chee · x · 2026-07-31
The author evaluated the agent capabilities of the latest DeepSeek-V4-Flash (248B parameters) within Codex CLI on a 3D/visual task to create a voxel pagoda garden.
Results:
- Cost & Efficiency: The run cost only $0.07 thanks to efficient token usage and good cache utilization.
- Time & Limitations: The process took 10 minutes, though about half the time was wasted because the gpt-image-2 skill required an OpenAI API key.
More from coding & agent
- Simon Willison: code mode is the future of tool calling, if it can be sandboxed — springrod · 2026-10-16
- Agent named 'Ilya' reportedly scores 1.1pt higher on benchmarks — daniel_mac8 · 2026-10-16
- Gemini CLI fix: substring matching inlined unrequested binary files into model context — priyanshu1976 · 2026-10-16
- Building AI agents on real healthcare claims data: the domain work that makes them work — alex_verem · 2026-10-16
- OpenAI stealth-testing mysterious internal model 'Aeon' in Codex desktop app and Dots — cedric_chee · 2026-10-16
- Vibe-coded iPhone Duo sniping bot built with Codex in 17 minutes, 41 tests passing — xiaohu · 2026-10-16