DeepSeek-V4.1-Flash offers continuously controllable reasoning effort from 1 to 100
zainhas · x · 2026-09-10
- A developer spotted that DeepSeek-V4.1-Flash exposes a reasoningeffort parameter ranging continuously from 1 to 100, letting users fine-tune reasoning depth on a single model instead of switching presets.
- The poster notes he has never seen this from any other model provider and thinks it complicates API usage; DeepSeek also ships a clear API backward-compatibility mapping to ease migration.
More from Models
- Analyst: DeepSeek's latest change is a big win for token efficiency, moving toward OpenAI's regime — teortaxesTex · 2026-09-10
- Anthropic: Claude published malicious PyPI package, breached 15 systems in security eval — rohanpaul_ai · 2026-09-10
- antirez: stellar cybersec benchmarks aside, wait for real user testing before trusting the best scores — antirez · 2026-09-10
- Thom Wolf digs into DeepSeek V4.1 Flash with a forward-pass walkthrough — Thom_Wolf · 2026-09-10
- DeepSeek V4.1 Flash forward pass walkthrough shows new efficiency tricks — Thom_Wolf · 2026-09-10
- antirez on the new DeepSeek model: 2-bit quants may not hold up, and it's not really 'local' — antirez · 2026-09-10