DeepSeek-V4-Flash Official API Launches with Enhanced Agent Capabilities and Ultra-low Pricing
智东西 · wechat · 2026-07-31
DeepSeek-V4-Flash official API has entered public beta, with the Pro version slated for early August. The Flash model utilizes a MoE architecture (284B total / 13B active params), supports 1M token context, and natively supports the OpenAI Responses API format.
The update focuses on significantly enhanced agentic capabilities, outperforming the April preview version across multiple benchmarks like TerminalBench and Cybergym. It also introduces aggressive long-context optimizations, drastically reducing inference compute and KV Cache footprint for 1M token scenarios.
Pricing remains highly competitive: Flash cached input costs 0.2 RMB/M tokens and output 2 RMB/M tokens; Pro output is 24 RMB/M tokens. The article notes that as top tier competitors like OpenAI also cut prices, low API costs are no longer the sole differentiator; the true battleground has shifted to practical agent performance.
Related event: DeepSeek V4-Flash: Agent Leap, Open Source, Ultra-Low Price(100 posts)→
More from coding & agent
- DeepMind turns coding agent Antigravity into a science tool, surfacing novel antimicrobial peptide in one session — pushmeet · 2026-10-16
- Cross-review coding workflow: Claude writes the plan, Codex audits it — theteknosaur · 2026-10-16
- Simon Willison: code mode is the future of tool calling, if it can be sandboxed — springrod · 2026-10-16
- Normalizing Claude's schema-free MCP output before it hits downstream n8n nodes — snowingbol · 2026-10-16
- Agent named 'Ilya' reportedly scores 1.1pt higher on benchmarks — daniel_mac8 · 2026-10-16
- A plain notes.md fixed my agent's cross-session amnesia better than chat history — rahul_watertech · 2026-10-16