Rumor: DeepSeek V4 Imminent with Peak/Off-Peak Pricing
新智元 · wechat · 2026-07-19
Rumors suggest the official release of DeepSeek V4 is imminent, with some developers already gaining early access. The new version is split into V4 Flash and V4 Pro, and its chain-of-thought opening phrasing has shifted from "Let me" to "I'm/I'll".
According to early testers, V4's overall capabilities approach Opus 4.8 levels, its coding performance rivals GPT-5.6 Sol, and its Agent and 3D/SVG generation capabilities have seen significant improvements. Regarding pricing, DeepSeek will introduce a peak/off-peak billing model for the first time: V4 Pro costs $1.74 per million output tokens during peak hours and $0.87 off-peak; V4 Flash is just $0.28 per million output tokens. Its cost-to-performance ratio remains exceptionally high compared to competitors.
More from Models
- Is DeepSeek's rumored K3 a scaled-down model, or something bigger? X users debate — teortaxesTex · 2026-09-11
- DeepSeek update keeps cache hits mid-conversation, cuts costs 36.6% — teortaxesTex · 2026-09-11
- AI Sextet offers 6 models free and unlimited for 14 days, including DeepSeek and Qwen — airesearch12 · 2026-09-11
- BullshitBench update: GPT-6-Astra beats all prior OpenAI models but still trails Anthropic — scaling01 · 2026-09-11
- Astra Scores 83% on GauntletBench, First Computer-Use Agent to Beat Human Baseline — ducha_aiki · 2026-09-11
- Kimi K2.8 Preview rolls out: near-K3 coding performance, 1M context for all tiers — teortaxesTex · 2026-09-11