🎉 DEEPSEEK-V4 JUST GOT MORE EXPENSIVE, BUT B.AI IS MAKING IT FREE!
DeepSeek-V4-Flash is now available on B.AI with unlimited, free access for a limited time! Fully supported across both Web Chat and API, whether you're running code, orchestrating AI Agents, processing long contexts, or executing high-concurrency workflows.
📋 How to Access
🌐 Web Chat: log in and select DeepSeek-V4-Flash from the model picker
🔧 API: go to the API Dashboard, choose Official Mode → Full Access, generate your API Key, copy the Model ID, start building
🧠 Breaking Down the Word "Just" in This Headline
Worth reading "just got more expensive" precisely, given the actual scale of DeepSeek's own price change, up to 1,100% depending on model, token type, and time of use, with a new peak/off-peak billing structure attached. "Just got more expensive" somewhat undersells how dramatic that specific increase actually was, worth being clear this isn't a modest price adjustment being offset, it's a substantial repricing being fully absorbed.
💭 My Take
That framing gap is worth flagging directly, because it changes how significant B.AI's response actually looks. A modest price bump offset by a free tier is a nice, routine promotional gesture. A 1,100%-in-places price hike fully offset by unlimited free access is a genuinely large absorption of cost on B.AI's part, worth recognizing the scale of what's actually being covered here rather than reading "just got more expensive" as understating a much bigger shift.
"Unlimited" continues to be the strongest specific claim in this offer, worth verifying directly once using it whether any practical rate limits show up despite the framing, the same open question worth checking against every prior version of this same announcement.
🎯 What I'd Actually Check
Whether "limited time" has a specific, stated end date anywhere in the official documentation, rather than remaining an open-ended qualifier across every version of this announcement so far.
🔗 Try it free: https://chat.b.ai/chat
@justinsuntron #TRONEcoStar
DeepSeek-V4-Flash is now available on B.AI with unlimited, free access for a limited time! Fully supported across both Web Chat and API, whether you're running code, orchestrating AI Agents, processing long contexts, or executing high-concurrency workflows.
📋 How to Access
🌐 Web Chat: log in and select DeepSeek-V4-Flash from the model picker
🔧 API: go to the API Dashboard, choose Official Mode → Full Access, generate your API Key, copy the Model ID, start building
🧠 Breaking Down the Word "Just" in This Headline
Worth reading "just got more expensive" precisely, given the actual scale of DeepSeek's own price change, up to 1,100% depending on model, token type, and time of use, with a new peak/off-peak billing structure attached. "Just got more expensive" somewhat undersells how dramatic that specific increase actually was, worth being clear this isn't a modest price adjustment being offset, it's a substantial repricing being fully absorbed.
💭 My Take
That framing gap is worth flagging directly, because it changes how significant B.AI's response actually looks. A modest price bump offset by a free tier is a nice, routine promotional gesture. A 1,100%-in-places price hike fully offset by unlimited free access is a genuinely large absorption of cost on B.AI's part, worth recognizing the scale of what's actually being covered here rather than reading "just got more expensive" as understating a much bigger shift.
"Unlimited" continues to be the strongest specific claim in this offer, worth verifying directly once using it whether any practical rate limits show up despite the framing, the same open question worth checking against every prior version of this same announcement.
🎯 What I'd Actually Check
Whether "limited time" has a specific, stated end date anywhere in the official documentation, rather than remaining an open-ended qualifier across every version of this announcement so far.
🔗 Try it free: https://chat.b.ai/chat
@justinsuntron #TRONEcoStar