{"slug": "deepseek-ai-deepseek-v4-flash-0731", "title": "deepseek-ai/DeepSeek-V4-Flash-0731", "summary": "DeepSeek released DeepSeek-V4-Flash-0731, a model that Artificial Analysis ranks ahead of MiniMax M3, a 428B model, and at $0.14 per million input tokens and $0.27 per million output tokens it may be the best value-per-intelligence model available. Simon Willison found that using the default reasoning level via OpenRouter produced disappointing results, but increasing the reasoning effort to high yielded much better output.", "body_md": "[deepseek-ai/DeepSeek-V4-Flash-0731](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731)\n\nArtificial Analysis [rank it](https://artificialanalysis.ai/models/deepseek-v4-flash) ahead of MiniMax M3 - a 428B model. It's $0.14/million input and $0.27/million output pricing means this may currently be the best value-per-intelligence model out there. It's looking very good on the [Intelligence Index vs. Cost per Intelligence Index Task](https://artificialanalysis.ai/models/deepseek-v4-flash#intelligence-comparison-tabs) chart:\n\nI got [a disappointing pelican](https://gist.github.com/simonw/83bfb1171792f1e7a4d8935b5e82317e#prompt) from it using the default reasoning level via OpenRouter:\n\nBut when I bumped reasoning level up to high I got [something much better](https://gist.github.com/simonw/83bfb1171792f1e7a4d8935b5e82317e#options):\n\n`llm -m openrouter/deepseek/deepseek-v4-flash-0731 -t pelican -o reasoning_effort high`\n\nVia [Hacker News](https://news.ycombinator.com/item?id=49120299)\n\nTags: [ai](https://simonwillison.net/tags/ai), [generative-ai](https://simonwillison.net/tags/generative-ai), [llms](https://simonwillison.net/tags/llms), [pelican-riding-a-bicycle](https://simonwillison.net/tags/pelican-riding-a-bicycle), [deepseek](https://simonwillison.net/tags/deepseek), [llm-release](https://simonwillison.net/tags/llm-release), [openrouter](https://simonwillison.net/tags/openrouter), [ai-in-china](https://simonwillison.net/tags/ai-in-china), [artificial-analysis](https://simonwillison.net/tags/artificial-analysis)", "url": "https://wpnews.pro/news/deepseek-ai-deepseek-v4-flash-0731", "canonical_source": "https://simonwillison.net/2026/Jul/31/deepseek-v4-flash-0731/#atom-everything", "published_at": "2026-07-31 23:59:44+00:00", "updated_at": "2026-08-01 00:31:58.145087+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-products", "generative-ai"], "entities": ["DeepSeek", "DeepSeek-V4-Flash-0731", "Artificial Analysis", "MiniMax M3", "Simon Willison", "OpenRouter", "Hacker News"], "alternates": {"html": "https://wpnews.pro/news/deepseek-ai-deepseek-v4-flash-0731", "markdown": "https://wpnews.pro/news/deepseek-ai-deepseek-v4-flash-0731.md", "text": "https://wpnews.pro/news/deepseek-ai-deepseek-v4-flash-0731.txt", "jsonld": "https://wpnews.pro/news/deepseek-ai-deepseek-v4-flash-0731.jsonld"}}