Hands-on: DeepSeek V4.1 Flash is fast at computer control, beats Codex for the author
AlchainHust · x · 2026-09-10
AlchainHust shares a hands-on with DeepSeek V4.1 Flash: not SOTA — it still struggles with tricky bug fixing — but extremely fast, and with max effort enabled it's good at breaking tasks into steps.
Combined with his own huashu-mac-use and huashu-chrome tools, it controls the Mac and browser quickly and stably, and in that scenario he finds it noticeably better than Codex and Claude Code.
More from coding & agent
- Agent verified 'success' after double-voting: a lesson in harness vs model bugs — TejasKumar_ · 2026-10-26
- diginomica: agentic automation hype is overdone, and 'The Hackening' of enterprise security has begun — jonerp · 2026-10-26
- 'Vibe verification': ~60% of AI-written formal specs prove the wrong property — basedjensen · 2026-10-26
- Devin launches Security suite powered by Claude Mythos with $100M+ credits — DevinAI · 2026-10-26
- HackBench injects reward hacks into agent trajectories to stress-test monitors — DanielKhashabi · 2026-10-26
- Hugging Face ships hf CLI updates: deduped cache saves 19GB, faster buckets, retryable GPU jobs — yb2698 · 2026-10-26