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

DeepSeek V4 Pro 0813

openrouter.ai

Post reply on HN
Print view View original

301–310 of 493 posts

Re: DeepSeek V4 Pro 0813

#301

Post by grono »

Lalabadie wrote: post #290
p1necone wrote: post #279

Earlier quoted context omitted.

50% cache hit is really low - in a standard agentic loop you should expect like 99%+ cache hit percentage (which should also lower that $12.50 to like a couple of $ for the same amount of tokens). If you're using a customised harness you should make sure you don't have something that's e.g. changing your system prompt on some requests or rewriting history - it can be tempting to do stuff like strip old thinking token…

In my experience, that's the OpenRouter tax. Even a session that does everything right to remain sticky ends up getting moved between providers on a few requests, which bills you the full context as input every time the switch happens. I assume it's done as load balancing/latency mitigation, but it's put me off of OpenRouter for my use cases (limited use, limited need for changing models).

Why not pin to specific openrouter provider and disable fallback?

Re: DeepSeek V4 Pro 0813

#302

Post by Lerc »

p1necone wrote: post #281
simonw wrote: post #153

Earlier quoted context omitted.

Default effort for OpenRouter. I'll try a grid of efforts... Wow, the low, medium, and high pelicans came out in surprisingly different styles: https://tools.simonwillison.net/markdown-svg-renderer#url=ht...

It's interesting that all three of those used roughly the same amount of tokens, and almost entirely output. Feels like the thinking level lever didn't alter cost at all for this specific task, even though it did change the output.

That raises the question of what is it actually doing?

If it isn't spending tokens on quality, is it the assumptions about the task difficulty that cause it to perform better? Or are their broader differences in the model being run.

Re: DeepSeek V4 Pro 0813

#303

Post by killingtime74 »

Gecko4072 wrote: post #10

Currently burning money quickly on official deepseek api. They are also increasing pricing starting today. V4 Flash 0731 still feels like the most outstanding model of the past few months and probably to come.

Just use opencode go, you get more bang for your buck. Same api

Re: DeepSeek V4 Pro 0813

#304

Post by Barbing »

benjiro29 wrote: post #176
NooneAtAll3 wrote: post #98

Earlier quoted context omitted.

I thought it was impossible to downvote posts?

I thought it was impossible to downvote posts? User Posts can be downvoted but you need over 500 karma to have access to the downvote button. A Submission can not be downvoted.

Yes and-

Submissions can be flagged by anyone and mods/admins can downweight them. (If I’m not mistaken this is common for, say, Flock posts at the moment.)

Curiosity & repetition are two key factors.

Re: DeepSeek V4 Pro 0813

#305

Post by julianz »

Lalabadie wrote: post #290
p1necone wrote: post #279

Earlier quoted context omitted.

50% cache hit is really low - in a standard agentic loop you should expect like 99%+ cache hit percentage (which should also lower that $12.50 to like a couple of $ for the same amount of tokens). If you're using a customised harness you should make sure you don't have something that's e.g. changing your system prompt on some requests or rewriting history - it can be tempting to do stuff like strip old thinking token…

In my experience, that's the OpenRouter tax. Even a session that does everything right to remain sticky ends up getting moved between providers on a few requests, which bills you the full context as input every time the switch happens. I assume it's done as load balancing/latency mitigation, but it's put me off of OpenRouter for my use cases (limited use, limited need for changing models).

There is only one provider for this model, so shouldn't be running into that.

Re: DeepSeek V4 Pro 0813

#306

Post by shunia_huang »

kees99 wrote: post #260
jklmnopqrstuvw wrote: post #211

Earlier quoted context omitted.

Months ago I start making this kind of test for my own reference. At beginning I I test each model multiple times, and results always same(pass or fail). Later I test only once for new models, I trust the results.

> multiple times, and results always same Not my experience at all. With smaller models, whenever I see a response that is going into wrong direction, I would just redo that step, and more often that not that brings improvement. This effect is less pronounced with SOTA, but still there.

Yes not my experience either.

I've tried or sometimes be stupid to work on bugs/features and ask with almost identical prompts with same modal and harness set, and yes, they generate totally different results.

Sometimes the output is unusable and even with extended guidance it will still drift away from what I was expecting.

Sometimes the output is just one shot and follows almost whatever I want.

I then be used to work like this, if the model and harness set does not work for one time, I just start a new session and do it again. And currently there is one of my task working like this.

Re: DeepSeek V4 Pro 0813

#308

Post by ticoombs »

khimaros wrote: post #280
simonw wrote: post #136

Nice bicycle chain, the little basket with a fish didn't show up in the right place: https://tools.simonwillison.net/markdown-svg-renderer#url=ht...

these links never work for me. always "Error: Enter a valid URL" when opening in Firefox. maybe a URL escape issue with Glider?

Can confirm. Also use glider which seems to double encode. I have to open the comment in Firefox/browser and then click on the link

Re: DeepSeek V4 Pro 0813

#309

Post by shunia_huang »

freakynit wrote: post #108

Just tested through openrouter.. gave exactly same task.. the task was to scan existing repo, and generate a single docker-compose file to deploy behind a caddy server, where certain port ranges are already used, the service demands widlcard certificates to be provisioned from outside, and postgre needs to be built-in one... Tested this model, and gpt-5.6-terra-high. Results: this one had few issues. terra: none. The…

The flash model will always use an outdated Treafik version that is not compatible with the newer docker engine, I tried to deploy some personal services with Traefik and everytime it uses this wrong version, and then fixes the version issue in the thinking chain.

I was thinking to switch to Caddy but with your experience I'm gonna stay with Traefik and bare with the version issue...

Re: DeepSeek V4 Pro 0813

#310

Post by blackqueeriroh »

Art9681 wrote: post #227
derangedHorse wrote: post #126

Earlier quoted context omitted.

Terra has not been able to do any of the technical tasks I've asked of it correctly. I'm surprised others get use out of it. Anything below Sol high tends to give me mostly unreliable results. I'm using codex as my main harness but maybe it performs better with a different one.

Terra is great. It's wild how different our experiences are. Install the Superpowers plugin. Behold.

Man, Superpowers is sooooo good. I use it with Opus, Sonnet, and Fable, and its review cycles are truly clutch.

I layer on CodeRabbit for PR Review and it’s just so solid.

Post reply on HN