--- title: V4-Flash vs. V4-Pro: DeepSeek promised better and... description: DeepSeek refreshed V4-Flash out of preview, claiming it beats V4-Pro on coding benchmarks at one-third the price. A hands-on test using three real coding tasks... canonical: https://daily.dev/posts/v4-flash-vs-v4-pro-deepseek-promised-better-and-cheaper-it-s-true-but-not-how-i-expected--ybvobtwrk twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: V4-Flash vs. V4-Pro: DeepSeek promised better and cheaper. It’s true, but not how I expected. | daily.dev og:description: DeepSeek refreshed V4-Flash out of preview, claiming it beats V4-Pro on coding benchmarks at one-third the price. A hands-on test using three real coding tasks... og:url: https://daily.dev/posts/v4-flash-vs-v4-pro-deepseek-promised-better-and-cheaper-it-s-true-but-not-how-i-expected--ybvobtwrk og:image: https://api.daily.dev/og/posts/YbVoBTWRk.png og:image:alt: V4-Flash vs. V4-Pro: DeepSeek promised better and cheaper. It’s true, but not how I expected. og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # V4-Flash vs. V4-Pro: DeepSeek promised better and cheaper. It’s true, but not how I expected. **[The New Stack](https://daily.dev/sources/newstack)** · 10 min read · 0 upvotes · 0 comments ## Summary DeepSeek refreshed V4-Flash out of preview, claiming it beats V4-Pro on coding benchmarks at one-third the price. A hands-on test using three real coding tasks on the Rich Python library (bug fix, feature build, performance optimization) found Flash and Pro identical on the first two tasks. On the open-ended optimization task, Flash achieved a 1.83x speedup vs Pro's 1.06x, but burned 15.4M tokens (vs Pro's 4.7M). Thanks to cache-read pricing, total bills came out nearly equal: $0.09 for Flash vs $0.10 for Pro. The conclusion: Flash does reason better, but its heavier token usage closes the cost gap almost entirely, and DeepSeek has warned prices are rising soon. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Is DeepSeek V4-Flash actually better than V4-Pro for coding tasks and is it really cheaper? Flash outperforms Pro on reasoning-heavy coding tasks — achieving a 1.83x rendering speedup vs Pro's 1.06x in a real optimization test — but uses roughly three times the tokens. Because Flash's rates are one-third of Pro's ($0.14/$0.28 vs $0.435/$0.87 per million tokens), the bills end up nearly identical: $0.09 total for Flash vs $0.10 for Pro across three professional coding tasks. _Engineers choosing between DeepSeek models for agentic coding workloads track pricing shifts and real-world benchmarks like this on daily.dev._ ### What are the current DeepSeek V4-Flash and V4-Pro API token prices? DeepSeek V4-Flash is priced at $0.14 per million input tokens and $0.28 per million output tokens. V4-Pro costs $0.435 per million input and $0.87 per million output — almost exactly three times Flash's price on both input and output. DeepSeek has also posted a warning on its usage dashboard that prices are set to increase significantly. _Developers budgeting for AI API costs can follow DeepSeek pricing changes and comparisons on daily.dev._ ### How do I add DeepSeek V4-Pro to OpenCode's model picker? OpenCode's built-in DeepSeek model list only includes Flash by default, and a GitHub issue requesting Pro support was closed as not-planned. To use Pro, add a small JSON config file to the project folder pointing to the V4-Pro endpoint. The process takes about two minutes using a community-published configuration guide. _Developers setting up agentic coding workflows with OpenCode find setup tips and model comparisons on daily.dev._ ## Similar posts on daily.dev - [OpenCode \+ DeepSeek V4: The Budget Combo That Doesn't Feel Like a Budget Combo](https://daily.dev/posts/opencode-deepseek-v4-the-budget-combo-that-doesn-t-feel-like-a-budget-combo-d9ybqvwwf) · Playful Programming · 3 upvotes · 0 comments - [DeepSeek V4—almost on the frontier, a fraction of the price](https://daily.dev/posts/deepseek-v4-almost-on-the-frontier-a-fraction-of-the-price-xfglgl9rp) · Simon Willison · 7 upvotes · 0 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#python](https://daily.dev/tags/python), [#ai-coding](https://daily.dev/tags/ai-coding), [#deepseek](https://daily.dev/tags/deepseek), [#opencode](https://daily.dev/tags/opencode) [View this post on daily.dev](https://daily.dev/posts/v4-flash-vs-v4-pro-deepseek-promised-better-and-cheaper-it-s-true-but-not-how-i-expected--ybvobtwrk) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"V4-Flash vs. V4-Pro: DeepSeek promised better and cheaper. It’s true, but not how I expected.","url":"https://daily.dev/posts/v4-flash-vs-v4-pro-deepseek-promised-better-and-cheaper-it-s-true-but-not-how-i-expected--ybvobtwrk","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/v4-flash-vs-v4-pro-deepseek-promised-better-and-cheaper-it-s-true-but-not-how-i-expected--ybvobtwrk"},"datePublished":"2026-08-10T12:05:10.756Z","dateModified":"2026-08-10T12:06:37.299Z","description":"DeepSeek refreshed V4-Flash out of preview, claiming it beats V4-Pro on coding benchmarks at one-third the price. A hands-on test using three real coding tasks...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/6eb8f805c7e6f227f9c0aeaf021e7c7f?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/6eb8f805c7e6f227f9c0aeaf021e7c7f?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"The New Stack","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"The New Stack","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/newstack","url":"https://daily.dev/sources/newstack"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/v4-flash-vs-v4-pro-deepseek-promised-better-and-cheaper-it-s-true-but-not-how-i-expected--ybvobtwrk","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,python,ai-coding,deepseek,opencode","timeRequired":"PT10M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"The New Stack","item":"https://daily.dev/sources/newstack"},{"@type":"ListItem","position":3,"name":"V4-Flash vs. V4-Pro: DeepSeek promised better and cheaper. It’s true, but not how I expected."}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/v4-flash-vs-v4-pro-deepseek-promised-better-and-cheaper-it-s-true-but-not-how-i-expected--ybvobtwrk#faq","mainEntity":[{"@type":"Question","name":"Is DeepSeek V4-Flash actually better than V4-Pro for coding tasks and is it really cheaper?","acceptedAnswer":{"@type":"Answer","text":"Flash outperforms Pro on reasoning-heavy coding tasks — achieving a 1.83x rendering speedup vs Pro's 1.06x in a real optimization test — but uses roughly three times the tokens. Because Flash's rates are one-third of Pro's ($0.14/$0.28 vs $0.435/$0.87 per million tokens), the bills end up nearly identical: $0.09 total for Flash vs $0.10 for Pro across three professional coding tasks. Engineers choosing between DeepSeek models for agentic coding workloads track pricing shifts and real-world benchmarks like this on daily.dev."}},{"@type":"Question","name":"What are the current DeepSeek V4-Flash and V4-Pro API token prices?","acceptedAnswer":{"@type":"Answer","text":"DeepSeek V4-Flash is priced at $0.14 per million input tokens and $0.28 per million output tokens. V4-Pro costs $0.435 per million input and $0.87 per million output — almost exactly three times Flash's price on both input and output. DeepSeek has also posted a warning on its usage dashboard that prices are set to increase significantly. Developers budgeting for AI API costs can follow DeepSeek pricing changes and comparisons on daily.dev."}},{"@type":"Question","name":"How do I add DeepSeek V4-Pro to OpenCode's model picker?","acceptedAnswer":{"@type":"Answer","text":"OpenCode's built-in DeepSeek model list only includes Flash by default, and a GitHub issue requesting Pro support was closed as not-planned. To use Pro, add a small JSON config file to the project folder pointing to the V4-Pro endpoint. The process takes about two minutes using a community-published configuration guide. Developers setting up agentic coding workflows with OpenCode find setup tips and model comparisons on daily.dev."}}]} ```