Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

news.ycombinator.com

Post reply on HN
Print view View original

51–60 of 216 posts

Re: DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

#51

Post by tjwebbnorfolk »

simonw wrote: post #49

> all requests to the Pro model will be routed to V4.1 Flash and billed at Flash's price If I'd carefully tested and optimized prompts against Pro I wouldn't be keen on this particular news. I feel like API model providers should lean towards not swapping out models on their paying customers, no matter how much "better" the new model is meant to be.

Sure, but if a company decided to place a remote chinese hedge fund's API at the center of a critical business workflow, this is a lesson better learned sooner rather than later.

Re: DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

#52

Post by tinyhouse »

jiehong wrote: post #20

Sounds nice! But, the web ui chat version of flash has very poor language following abilities in my experience: You may ask it something in English, and get a thinking chain in Chinese with an answer in Chinese, or an English thinking chain and an English answer. Using the retry button on the same question has a 50/50 chance of any of those results. Sometimes, asking something in English, but where information are mo…

I've been working with Pro and it's been great so far.

Re: DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

#53

Post by nicce »

serf wrote: post #41
EbNar wrote: post #15

Since a few months, I almost exclusively use the Chinese "flash" models for my needs. They are a joy and they cost pennies per answer. Great job.

I recently had to config my harness to watch for cybersecurity flags from astra and funnel requests to flash when they occur because Astra gets queezy when you talk to it about UDP packets in games. Works fantastic. Glad there is a more 'uncensored' thing to fall back to when the frontier folk are too sensitive.

The only positive side is that it is harder for students to feed university exercises to the agent in cybersecurity and expect it to make them all.

Re: DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

#54

Post by KyleTheDev »

hinow wrote: post #28

But what no one mentions is that the price is going from a starting point of $0.16 to $0.60, so basically they're charging nearly four times as much.

Like _aavaa_ said, make sure you're comparing the right vals 1:1. There's different costs for cache hit, cache misses, output tokens, etc. This one seems, during non-peak hours, cheaper. Peak hours are obviously more expensive, if they're gonna be 2x non-peak pricing. But, that might end up decreasing in the future.

Re: DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

#55

Post by KyleTheDev »

HansHamster wrote: post #36
calgoo wrote: post #17

Earlier quoted context omitted.

v4 flash has been working quite well for the majority of my personal projects, with occasional v4 pro or Kimi 3 for the most complicated tasks or to check the overall project progress (when vibe coding).

I must be doing something wrong. I gave v4 pro a try a couple of days ago, gave it a simple prompt like "clean up functions x and y in file z" and it would always start off promising, just to quickly get sidetracked, start hallucinating problems in the code, and just get stuck for hours until I interrupt it: — hmm — 0x2D696370 — little-endian bytes: 70 63 69 2D = 'p','c','i','-' — hmm — WAIT — WAIT — !!!!! — *WAIT —…

I've been using the 0731 Flash V4 model, via Opencode, and I've had no major issues. It feels very comparable to Opus 4.6/4.7, that I use at work. I haven't ran into any of the problem you mention, so that might be a quirk of V4 pro, the specific harness, or maybe the host you're using?

Re: DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

#56

Post by edude03 »

I've been watching a bunch of bycloud on YouTube recently, and although he's done a great job reviewing papers from the big AI labs, I feel like I'm missing something - how have all the labs seemingly made a model that's cheaper, faster AND has better performance? Historically `flash` variants (like codex spark as well) have been faster but perform worse

Re: DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

#57

Post by mattmcal »

jiehong wrote: post #20

Sounds nice! But, the web ui chat version of flash has very poor language following abilities in my experience: You may ask it something in English, and get a thinking chain in Chinese with an answer in Chinese, or an English thinking chain and an English answer. Using the retry button on the same question has a 50/50 chance of any of those results. Sometimes, asking something in English, but where information are mo…

It's not just web chat, V4 Flash 7/31 suffers from a lot of pathological behavior in coding harnesses as well, e.g. infinite loops, hallucinations, premature termination, and invalid tool calls.

Re: DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

#58

Post by kgeist »

el_io wrote: post #26
jiehong wrote: post #20

Sounds nice! But, the web ui chat version of flash has very poor language following abilities in my experience: You may ask it something in English, and get a thinking chain in Chinese with an answer in Chinese, or an English thinking chain and an English answer. Using the retry button on the same question has a 50/50 chance of any of those results. Sometimes, asking something in English, but where information are mo…

I'm also totally not sure why it do that, but I guess because they're searching from China and web results comeback in Chinese so the model start using that.

The web UI's system prompt is also probably in Chinese

Re: DeepSeek launching v4.1 flash cheaper and more capable than v4 pro

#60

Post by _3u10 »

edude03 wrote: post #56

I've been watching a bunch of bycloud on YouTube recently, and although he's done a great job reviewing papers from the big AI labs, I feel like I'm missing something - how have all the labs seemingly made a model that's cheaper, faster AND has better performance? Historically `flash` variants (like codex spark as well) have been faster but perform worse

That’s how increasing performance works. You make a model 10x faster, then you make it think 2x as much.

Its cost is now 1/10th per token, and 1/5th per task.

Basically they have shitty hardware so they have to do a lot of optimization. Think of it like replacing an O(n) algorithm with O(log n).

Anthropic / Open AI think the best path is the most intelligent models deepseek is more focused on tok/$

Post reply on HN