DeepSeek ships V4-Flash-0731 on a Saturday, and the frontier drop stops being an event
DeepSeek-V4-Flash-0731, dated to its release day of 31 July, lands into a tracker already counting 335+ frontier model releases across major labs this cycle. A year ago a new frontier model was a keynote. Now it is a filename with a date suffix, shipped on a weekend, one of several that week.
The Flash designation is the tell. DeepSeek is no longer racing to top a single leaderboard with one flagship; it is shipping a family — a Pro for capability, a Flash for latency and cost — and iterating the whole line on a weekly cadence. The 0731 date-stamp in the name is an admission that the model is a checkpoint, not a monument, and that another will follow before the coverage of this one finishes.
The competitive frame is compression of the release cycle itself. Anthropic shipped Opus 5 on 24 July; OpenAI's GPT-5.6 line (Sol, Luna, Terra) is weeks old; xAI's Grok 4.5 is in the same window. When four labs each ship a frontier-class model inside a month, the marginal model stops moving the market and starts moving the benchmark by a point.
For buyers the practical consequence is that model choice is now a moving target measured in weeks, and any procurement that pins a single model version for a year is pinning to something that will be two generations stale by renewal. The durable decision is the abstraction layer — the gateway, the eval harness — not the model behind it.
LLM-Stats — AI updates today (August 2026) — latest model releases → · AI Release Tracker — Complete LLM timeline 2022-2026 →