DeepSeek V4 – almost on the frontier • Hacker News Board

Re: DeepSeek V4 – almost on the frontier

#81

Post by cheshire_cat » Sat, May 02, 2026, 11:55 AM UTC

cassianoleal wrote: post #75

cheshire_cat wrote: post #69

While the cost are lower than frontier models there are two factors that make DS4 Pro and K2.6 not as cheap as they might look. For DS4 Pro there's a discount going on for the official API, which sometimes gets overlooked and mixed up in discussions. Simon uses the full price in the comparison, so that's not an issue here. The other issue is that DS4 Pro and K2.6 often use way more reasoning tokens than the frontier…

Sure that can happen but it hasn’t been my experience. I just spent a whole day using it for some pretty hefty refactors, many rounds of back-and-forths, thousands of lines of code changes, reviews, investigations, many subagents running parallel tasks, the works. Total cost $0.95, altogether. I had attempted this with Opus 4.6 in the past and it burned through the $10 budget I’d given it before it returned from my i…

Sounds promising, thanks for your report.

I didn't want to say that they're not cheaper to run, artificial analysis also shows that they're cheaper. My main point was about it being important to also look at token efficiency, not only cost per token, to get the full picture.

Re: DeepSeek V4 – almost on the frontier

#82

Post by vdfs » Sat, May 02, 2026, 11:56 AM UTC

stavros wrote: post #47

amelius wrote: post #36

Earlier quoted context omitted.

Aren't they all billing at discount?

Anthropic's and OpenAI's costs seem to include a fairly ok margin, from the very fourth hand info I have.

In total, how many hands do you have?

Re: DeepSeek V4 – almost on the frontier

#83

Post by FrasiertheLion » Sat, May 02, 2026, 11:57 AM UTC

swiftcoder wrote: post #79

FrasiertheLion wrote: post #74

Earlier quoted context omitted.

Have you given GLM 5.1 or Kimi K2.6 a shot for coding? They outperform Deepseek v4 pro.

> Kimi K2.6 a shot for coding? They outperform Deepseek v4 pro I think this probably depends quite a bit on the specific problem. I'm finding that Deepseek v4 Flash often outdoes Kimi 2.6 on a variety of coding problems that involve complex spatial reasoning

Oh that's quite interesting and hasn't been my experience with regular backend code specifically with respect to tool calling. However that could be because the tool calling format in vllm for Deepseek v4 was broken until a few days ago and that's how I'm running it.

I've been hearing amazing things about Flash, I should give it a try.

Re: DeepSeek V4 – almost on the frontier

#84

Post by utopiah » Sat, May 02, 2026, 11:58 AM UTC

vdfs wrote: post #82

stavros wrote: post #47

Earlier quoted context omitted.

Anthropic's and OpenAI's costs seem to include a fairly ok margin, from the very fourth hand info I have.

In total, how many hands do you have?

Those aren't their hands.

Re: DeepSeek V4 – almost on the frontier

#85

Post by rane » Sat, May 02, 2026, 11:59 AM UTC

taffydavid wrote: post #21

I tried deepseek v4 through open code at the weekend. I'm a daily Claude/Claude code user. I tried to build something simple and while it got the job done the thinking displayed did not fill me with confidence. It was pages and pages of "actually no", "hang on", "wait that makes no sense". It was like the model was having a breakdown. Bear in mind open code was also new to me so I could be just seeing thinking where…

You can just use it through Claude Code, so you get to keep the system prompt and tooling you are used to.

3rd party models are a drop-in replacement with `ANTHROPIC_BASE_URL` in Claude Code, something people seem to miss right now. And contrary to what Anthropic might like to have you think, you don't need Opus 4.7 to run the harness to get similar performance.

https://api-docs.deepseek.com/quick_start/agent_integrations...

Re: DeepSeek V4 – almost on the frontier

#86

Post by gessha » Sat, May 02, 2026, 12:03 PM UTC

vdfs wrote: post #82

stavros wrote: post #47

Earlier quoted context omitted.

Anthropic's and OpenAI's costs seem to include a fairly ok margin, from the very fourth hand info I have.

In total, how many hands do you have?

Enough to reach the bottom of the rabbit hole.

Re: DeepSeek V4 – almost on the frontier

#87

Post by bilsbie » Sat, May 02, 2026, 12:03 PM UTC

Dumb question? Why does pro make a worse pelican than flash?

Re: DeepSeek V4 – almost on the frontier

#88

Post by Gracana » Sat, May 02, 2026, 12:06 PM UTC

holysantamaria wrote: post #16

From the pricing page of deepseek: (3) The deepseek-v4-pro model is currently offered at a 75% discount, extended until 2026/05/31 15:59 UTC. Was this taken into account when reviewing the model?

The article quotes the full price.

Re: DeepSeek V4 – almost on the frontier

#89

Post by girvo » Sat, May 02, 2026, 12:09 PM UTC

kay_o wrote: post #58

taffydavid wrote: post #21

I tried deepseek v4 through open code at the weekend. I'm a daily Claude/Claude code user. I tried to build something simple and while it got the job done the thinking displayed did not fill me with confidence. It was pages and pages of "actually no", "hang on", "wait that makes no sense". It was like the model was having a breakdown. Bear in mind open code was also new to me so I could be just seeing thinking where…

Before CC and Codex removed thinking/verbose and hid most of it, both do that .

Yeah people aren’t aware that we don’t see the actual traces anymore lol

Re: DeepSeek V4 – almost on the frontier

#90

Post by ghm2180 » Sat, May 02, 2026, 12:09 PM UTC

I've been using the planning framework from Matt Pocock on very typical brownfield code. I use a harness over claude code, this is so cheap that I would be tempted to mirror my initial prompt to it and compare their responses to the task.