DeepSeek-V4-Flash正式版API上线公测

DeepSeek于7月31日宣布DeepSeek-V4-Flash正式版API上线公测134。该版本在多项Agent基准测试中表现出色，Terminal Bench 2.1达到82.7分134，NL2Repo得分54.2分23，Cybergym达到76.7分234，DSBench-FullStack为68.7分134，DSBench-Hard为59.6分134。

DeepSeek-V4-Flash-0731与preview版本保持相同的模型结构和参数规模，仅通过重新后训练实现性能优化134。新版本原生支持Responses API格式并针对性适配Codex34。此次更新仅涉及V4-Flash API接口升级，V4-Pro API与应用/网页端模型暂未变更14。官方表示V4-Pro正式版将尽快发布34。

两个旧版API模型名称deepseek-chat和deepseek-reasoner将在三个月后停用1。

DeepSeek launched the official version of its DeepSeek-V4-Flash API into public testing on July 31st 134. The update marks a significant enhancement to the model's agent capabilities, with the company achieving substantially improved performance across multiple benchmark tests while maintaining the same model architecture and parameters as the preview version 134.

The V4-Flash official release demonstrates notable gains in Agent-focused tasks. The model achieved a score of 82.7 on Terminal Bench 2.1 134, 76.7 on Cybergym 34, 68.7 on DSBench-FullStack 13, and 59.6 on DSBench-Hard 13. Additional benchmark results include 54.2 on NL2Repo, 70.3 on Toolathlon verified, 54.4 on DeepSWE, 25.2 on Agent Last Exam, and 25.1 on Automation Bench (Public) 3. DeepSeek achieved these improvements through retraining optimization alone, without altering the model's underlying structure or parameter scale 134.

The V4-Flash official version natively supports the Responses API format and includes targeted adaptation for Codex 34. The upgrade applies exclusively to the V4-Flash API interface, while the V4-Pro API and application/web-based models remain unchanged 14. The company indicated that the V4-Pro official version will be released as soon as possible 34. Additionally, two legacy API model names, deepseek-chat and deepseek-reasoner, will be discontinued three months later 1.