DeepSeek 4.1 Flash run costs $0.0219 over 5 minutes with 93.6% cache rate
rickasaurus · x · 2026-11-18
A developer ran a 5 min 14 sec task on DeepSeek 4.1 Flash for just $0.0219: OpenRouter billed 1,264,512 cached and 86,612 uncached input tokens — a 93.6% cache rate. At that pace, inference would cost roughly $0.251/hour, or $181 per 30 days running nonstop. His verdict: it's basically free to run for personal use and an amazing model for the cost.
Related event: DeepSeek 4.1 Flash Tested: Near-Free for Long Agent Tasks(2 posts)→
More from Models
- Snap test: Claude admits it can't claim 'ZERO qualia' in that moment — Sauers_ · 2026-11-18
- Speculation Swirls That OpenAI's 6.2 Astra Leans on Looping Transformers, Raising Monitorability Concerns — scaling01 · 2026-11-18
- Could a GPT-6.2 Astra + Opus 5.5 fusion fix the model-human interface? — daniel_mac8 · 2026-11-18
- Blogger ditches Fable 5.2 over fast limit burnout, returns to Opus 5.5 — scaling01 · 2026-11-18
- Developer: DeepSeek 4.1 Flash is basically free to run for personal use — rickasaurus · 2026-11-18
- LLM reviewers draw 50% of weaknesses from the appendix vs 8% for humans — plopesresearch · 2026-11-18