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

DeepSeek V4 Flash 0731

arcprize.org

Post reply on HN
Print view View original

101–110 of 477 posts

Re: DeepSeek V4 Flash 0731

#101

Post by dan_q »

_s_a_m_ wrote: post #93
LaurensBER wrote: post #9

I've been using it extensively since the release and the best summary I can give is that it's good enough to use it for (almost) everything and cheap enough that the cost are irrelevant. I'm running it in Oh My Pi with a second instance running as "advisor" and even with 5-6 active sessions (effectively 12 streams) I'm struggling to spend more than 5 bucks per day. OpenCode Go even has double limits temporarily so fo…

These posts have to be Chinese bots, these models are all trash. Used it via OpenCode for an hour, cost me one hour of my life. It is for anything complete trash.

You're mad.

Re: DeepSeek V4 Flash 0731

#102

Post by dyauspitr »

542458 wrote: post #4

Kimi K3 was an interesting model only a month ago, and now we're looking at the same performance for 1/20th of the price. Wild how fast this is advancing.

Not for long, Deepseek is saying they will have a significant price jump soon. They really shouldn’t do it because they are on the cusp of capturing the scalable API market.

Re: DeepSeek V4 Flash 0731

#103

Post by vb-8448 »

modeless wrote: post #84

DeepSeek has announced an upcoming "significant increase" in price, so this line may have to move to the right soon. https://api-docs.deepseek.com/quick_start/pricing/

Why? It's open weight, there are plenty providers on open router that are serving the latest v4 flash at 0.14/0.28 $.

Re: DeepSeek V4 Flash 0731

#106

Post by greenavocado »

akman wrote: post #39
NorwegianDude wrote: post #32

Earlier quoted context omitted.

Eh, what are you guys even talking about? Deepseek is not cheapest provider as is, and it's MIT. So deepseek making it more expensive to use is just nonsense, they can only change their own pricing. It's the beauty of MIT license and open weights. If anything, these models are some of the safest in the world to use if you worry about a rug pull.

90%+ cache hit rate is common, and so you'll see on places like openrouter that Deepseek cache cost is indeed a magnitude cheaper than the rest.

My usage thus far from api.deepseek.com

  - input_cache_hit_tokens: 1,265,646,976 x 0.0000000028 = $3.5438115328
  - input_cache_miss_tokens: 18,208,088 x 0.00000014 = $2.54913232
  - output_tokens: 9,615,178 x 0.00000028 = $2.69224984
  - request_count: 10,837 (no price)
Total cost: $8.7851936928 (approximately $8.79)

Cache:

  - Hit: 1,265,646,976
  - Miss: 18,208,088
  - Total input tokens: 1,283,855,064
Hit rate: 98.582% (1,265,646,976 / 1,283,855,064)

Re: DeepSeek V4 Flash 0731

#107

Post by walrus01 »

Oke of the great advantages of v4 flash 0731 is that even in the largest size unsloth quantized gguf, Q8 K XL, it will fit well within the resources of a 256GB DRAM server. If you have no gpu at all and are okay with setting up a workflow that handles slow token per second rate, give it a task and check back in 4-6 hours, it works great. And remember to give it more lengthy tasks to run overnight. Whatever workflow you set up, the idea is to keep it busy 24x7 doing different things in parallel.

Re: DeepSeek V4 Flash 0731

#109

Post by alex0015 »

greenavocado wrote: post #96
_s_a_m_ wrote: post #93

Earlier quoted context omitted.

These posts have to be Chinese bots, these models are all trash. Used it via OpenCode for an hour, cost me one hour of my life. It is for anything complete trash.

(1) you used opencode (2) what provider did you use. openrouter is trash because they shit up the model serving. no max effort and horrific cache utilization, on the order of 50-75%, absolutely garbage. beware

What should we be running deepseek on besides opencode? I chose it because I heard good things. Also provider is directly through deepseek credits.

Re: DeepSeek V4 Flash 0731

#110

Post by SwellJoe »

eli wrote: post #81
SwellJoe wrote: post #71

DeepSeek is my cheap and cheerful Chinese model of choice for API use. Has been for a while, but now it's Flash instead of Pro. Even cheaper, and now better then Pro. I feel like most of the major Chinese models are benchmaxxed, they have weird quirks every time I use them (Qwen 3.8 Max doesn't check its work and leaves stuff broken, doesn't write tests unless prompted, etc., Kimi ends up being quite expensive and ra…

Qwen 3.8 Max is very strong at troubleshooting and code review.

I'll grant it's very thorough when assigned a troubleshooting task. I'm not as confident of its code review though it is very good at security vulnerability auditing, and isn't hobbled for that work like Fable, and even Opus refuses some work in that area now.
Post reply on HN