{"rewrite":{"id":"r_325deac26559da68a61c5b32","clusterId":"c_092573a37ea9bfcffc2d548d","slug":"deepseek-releases-v4-flash-0731-open-model","model":"deepseek-v4-flash:free","headline":"DeepSeek Releases V4-Flash-0731 Open Model","summary":"DeepSeek has released DeepSeek-V4-Flash-0731 as an open model under a commercial-use license. The MoE model has 284 billion total parameters and 13 billion active parameters. Third-party tests by Artificial Analysis rate its intelligence on par with Google's Gemini 3.6 Flash, and it outperforms GLM-5.2 in coding and GPT-5.6 Luna in agent performance.","whyItMatters":"The release continues DeepSeek's pattern of high-performance open models, with V4-Flash-0731 offering cost efficiency below the recently discounted GPT-5.6 Luna.","webCardHtml":"\u003cp\u003eDeepSeek has published DeepSeek-V4-Flash-0731 on Hugging Face and ModelScope under the MIT License, making it freely available for commercial use. The model is an update to April\u0026#39;s DeepSeek-V4-Flash and uses a mixture-of-experts architecture with 284 billion total parameters and 13 billion active parameters.\u003c/p\u003e\u003cp\u003eAccording to Artificial Analysis, the model matches Gemini 3.6 Flash in intelligence, trails only Kimi K3 and GLM-5.2 among open models, and beats GLM-5.2 in coding. Its agent performance exceeds Gemini 3.6 Flash by a wide margin and edges out GPT-5.6 Luna by 0.1 points. The model also stands out for its efficiency, costing less per task than the late-July discounted GPT-5.6 Luna while delivering comparable intelligence.\u003c/p\u003e","blueskyPost":"DeepSeek released V4-Flash-0731, an open MoE model with 284B total params. It matches Gemini 3.6 Flash in intelligence, beats GLM-5.2 in coding, and costs less than GPT-5.6 Luna. MIT license, free on Hugging Face and ModelScope.","twitterPost":"DeepSeek released V4-Flash-0731, an open MoE model (284B total, 13B active). Tests show it matches Gemini 3.6 Flash, beats GLM-5.2 in coding, and costs less than GPT-5.6 Luna. MIT license, free on Hugging Face and ModelScope.","threadsPost":null,"newsletterBlurb":"DeepSeek has released V4-Flash-0731, an open model with 284 billion total parameters, under the MIT License. Independent tests rate its intelligence on par with Gemini 3.6 Flash, and it outperforms GLM-5.2 in coding and GPT-5.6 Luna in agent performance while costing less per task.","attributionJson":"[{\"source\":\"GIGAZINE\",\"url\":\"https://gigazine.net/news/20260803-deepseek-v4-flash-0731/\",\"title\":\"「DeepSeek-V4-Flash-0731」が登場、GPT-5.6 Lunaより安価で同等性能のオープンモデル\"}]","lintFlagsJson":null,"lintHits":0,"costUsd":0,"inputTokens":5330,"outputTokens":677,"status":"published","repairAttempts":0,"nextRepairAt":null,"factsAttemptedAt":1786301646,"createdAt":"2026-08-09T18:49:44.000Z","publishedAt":"2026-08-09T18:51:44.000Z","updatedAt":"2026-08-09T18:49:44.000Z"},"cluster":{"id":"c_092573a37ea9bfcffc2d548d","canonicalTitle":"「DeepSeek-V4-Flash-0731」が登場、GPT-5.6 Lunaより安価で同等性能のオープンモデル","representativeArticleId":"a_79d3acd98d71e4cc2af53817","sourceCount":1,"writtenSourceCount":1,"writeAttempts":0,"isSolo":true,"entitiesJson":"{\"anime_titles\":[],\"manga_titles\":[],\"work_titles\":[\"DeepSeek-V4-Flash-0731\"],\"studios\":[\"DeepSeek\"],\"people\":[],\"type\":\"announcement\",\"domain\":\"other\",\"is_roundup\":false}","contentType":"news","status":"published","firstSeenAt":"2026-08-03T03:45:00.000Z","lastSeenAt":"2026-08-03T03:45:00.000Z","updatedAt":"2026-08-09T18:51:45.000Z"},"attribution":[{"source":"GIGAZINE","url":"https://gigazine.net/news/20260803-deepseek-v4-flash-0731/","title":"「DeepSeek-V4-Flash-0731」が登場、GPT-5.6 Lunaより安価で同等性能のオープンモデル"}],"entities":{"anime_titles":[],"manga_titles":[],"work_titles":["DeepSeek-V4-Flash-0731"],"studios":["DeepSeek"],"people":[],"type":"announcement","domain":"other","is_roundup":false},"keyFacts":null}