DeepSeek v4

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

327•impact_sy•2h ago

Comments

luyu_wu•1h ago

For those who didn't check the page yet, it just links to the API docs being updated with the upcoming models, not the actual model release.

talim•1h ago

Weights are on Huggingface FWIW. https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/tree/main

cmrdporcupine•1h ago

My submission here https://news.ycombinator.com/item?id=47885014 done at the same time was to the weights.

dang, probably the two should be merged and that be the link

culi•53m ago

there's no pinging. Someone's gotta email dang

seanobannon•1h ago

Weights available here: https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro

BoorishBears•4m ago

https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-Base https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-Base

And we got new base models, wonderful, truly wonderful

nthypes•1h ago

https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/main...

Model was released and it's amazing. Frontier level (better than Opus 4.6) at a fraction of the cost.

sergiotapia•1h ago

The dragon awakes yet again!

kindkang2024•32m ago

There appears a flight of dragons without heads. Good fortune.

That's literally what the I Ching calls "good fortune."

Competition, when no single dragon monopolizes the sky, brings fortune for all.

rapind•1h ago

Pop?

onchainintel•1h ago

How does it compare to Opus 4.7? I've been immersed in 4.7 all week participating in the Anthropic Opus 4.7 hackathon and it's pretty impressive even if it's ravenous from a token perspective compared to 4.6

greenknight•1h ago

The thing is, it doesnt need to beat 4.7. it just needs to do somewhat well against it.

This is free... as in you can download it, run it on your systems and finetune it to be the way you want it to be.

johnmaguire•1h ago

... if you have 800 GB of VRAM free.

inventor7777•45m ago

I remember reading about some new frameworks have been coming out to allow Macs to stream weights of huge models live from fast SSDs and produce quality output, albeit slowly. Apart from that...good luck finding that much available VRAM haha

p1esk•1h ago

Do you think a lot of people have “systems” to run a 1.6T model?

applfanboysbgon•39m ago

No, but businesses do. Being able to run quality LLMs without your business, or business's private information, being held at the mercy of another corp has a lot of value.

choldstare•30m ago

Not really - on prem llm hosting is extremely labor and capital intensive

applfanboysbgon•25m ago

But can be, and is, done. I work for a bootstrapped startup that hosts a DeepSeek v3 retrain on our own GPUs. We are highly profitable. We're certainly not the only ones in the space, as I'm personally aware of several other startups hosting their own GLM or DeepSeek models.

forrestthewoods•20m ago

What type of system is needed to self host this? How much would it cost?

p1esk•11m ago

Depends on fast you want it to be. I’m guessing a couple of $10k mac studio boxes could run it, but probably not fast enough to enjoy using it.

disiplus•6m ago

Depends how many users you have and what is "production grade" for you but like 500k gets you a 8x B200 machine.

kelseyfrog•58m ago

What's the hardware cost to running it?

slashdave•52m ago

"if you have to ask..."

redox99•46m ago

Probably like 100 USD/hour

bbor•18m ago

I was curious, and some [intrepid soul](https://wavespeed.ai/blog/posts/deepseek-v4-gpu-vram-require...) did an analysis. Assuming you do everything perfectly and take full advantage of the model's MoE sparsity, it would take:

- To run at full precision: "16–24 H100s", giving us ~$400-600k upfront, or $8-12/h from [us-east-1](https://intuitionlabs.ai/articles/h100-rental-prices-cloud-c...).

- To run with "heavy quantization" (16 bits -> 8): "8xH100", giving us $200K upfront and $4/h.

- To run truly "locally"--i.e. in a house instead of a data center--you'd need four 4090s, one of the most powerful consumer GPUs available. Even that would clock in around $15k for the cards alone and ~$0.22/h for the electricity (in the US).

Truly an insane industry. This is a good reminder of why datacenter capex from since 2023 has eclipsed the Manhattan Project, the Apollo program, and the US interstate system combined...

zargon•9m ago

That article is a total hallucination.

onchainintel•49m ago

Completely agree, not suggesting it needs ot just genuinely curious. Love that it can be run locally though. Open source LLMs punching back pretty hard against proprietary ones in the cloud lately in terms of performance.

rvz•1h ago

It is more than good enough and has effectively caught up with Opus 4.6 and GPT 5.4 according to the benchmarks.

It's about 2 months behind GPT 5.5 and Opus 4.7.

As long as it is cheap to run for the hosting providers and it is frontier level, it is a very competitive model and impressive against the others. I give it 2 years maximum for consumer hardware to run models that are 500B - 800B quantized on their machines.

It should be obvious now why Anthropic really doesn't want you to run local models on your machine.

colordrops•41m ago

What's going to change in 2 years that would allow users to run 500B-800B parameter models on consumer hardware?

DiscourseFan•29m ago

I think its just an estimate

snovv_crash•28m ago

With the ability of the Qwen3.6 27B, I think in 2 years consumers will be running models of this capability on current hardware.

deaux•21m ago

Vibes > Benchmarks. And it's all so task-specific. Gemini 3 has scored very well in benchmarks for very long but is poor at agentic usecases. A lot of people prefering Opus 4.6 to 4.7 for coding despite benchmarks, much more than I've seen before (4.5->4.6, 4->4.5).

Doesn't mean Deepseek v4 isn't great, just benchmarks alone aren't enough to tell.

doctoboggan•1h ago

Is it honestly better than Opus 4.6 or just benchmaxxed? Have you done any coding with an agent harness using it?

If its coding abilities are better than Claude Code with Opus 4.6 then I will definitely be switching to this model.

madagang•56m ago

Their Chinese announcement says that, based on internal employee testing, it is not as good as Opus 4.6 Thinking, but is slightly better than Opus 4.6 without Thinking enabled.

mchusma•50m ago

I appreciate this, makes me trust it more than benchmarks.

deaux•20m ago

That's super interesting, isn't Deepseek in China banned from using Anthropic models? Yet here they're comparing it in terms of internal employee testing.

NitpickLawyer•35m ago

> (better than Opus 4.6)

There we go again :) It seems we have a release each day claiming that. What's weird is that even deepseek doesn't claim it's better than opus w/ thinking. No idea why you'd say that but anyway.

Dsv3 was a good model. Not benchmaxxed at all, it was pretty stable where it was. Did well on tasks that were ood for benchmarks, even if it was behind SotA.

This seems to be similar. Behind SotA, but not by much, and at a much lower price. The big one is being served (by ds themselves now, more providers will come and we'll see the median price) at 1.74$ in / 3.48$ out / 0.14$ cache. Really cheap for what it offers.

The small one is at 0.14$ in / 0.28$ out / 0.028$ cache, which is pretty much "too cheap to matter". This will be what people can run realistically "at home", and should be a contender for things like haiku/gemini-flash, if it can deliver at those levels.

0xbadcafebee•34m ago

I don't think we need to compare models to Opus anymore. Opus users don't care about other models, as they're convinced Opus will be better forever. And non-Opus users don't want the expense, lock-in or limits.

As a non-Opus user, I'll continue to use the cheapest fastest models that get my job done, which (for me anyway) is still MiniMax M2.5. I occasionally try a newer, more expensive model, and I get the same results. I have a feeling we might all be getting swindled by the whole AI industry with benchmarks that just make it look like everything's improving.

bbor•32m ago

For the curious, I did some napkin math on their posted benchmarks and it racks up 20.1 percentage point difference across the 20 metrics where both were scored, for an average improvement of about 2% (non-pp). I really can't decide if that's mind blowing or boring?

Claude4.6 was almost 10pp better at at answering questions from long contexts ("corpuses" in CorpusQA and "multiround conversations" in MRCR), while DSv4 was a staggering 14pp better at one math challenge (IMOAnswerBench) and 12pp better at basic Q&A (SimpleQA-Verified).

Quasimarion•29m ago

FWIW it's also like 10x cheaper.

taosx•1h ago

MErge? https://news.ycombinator.com/item?id=47885014

gbnwl•1h ago

I’m deeply interested and invested in the field but I could really use a support group for people burnt out from trying to keep up with everything. I feel like we’ve already long since passed the point where we need AI to help us keep up with advancements in AI.

wordpad•1h ago

The players barely ever change. People don't have problems following sports, you shouldn't struggle so much with this once you accept top spot changes.

ehnto•31m ago

It is funny seeing people ping pong between Anthropic and ChatGPT, with similar rhetoric in both directions.

At this point I would just pick the one who's "ethics" and user experience you prefer. The difference in performance between these releases has had no impact on the meaningful work one can do with them, unless perhaps they are on the fringes in some domain.

Personally I am trying out the open models cloud hosted, since I am not interested in being rug pulled by the big two providers. They have come a long way, and for all the work I actually trust to an LLM they seem to be sufficient.

DiscourseFan•26m ago

I find ChatGPT annoying mostly

awakeasleep•23m ago

Open settings > personalization. Set it to efficient base style. Turn off enthusiasm and warmth. You’re welcome

gbnwl•9m ago

I didn't express this well but my interest isn't "who is in the top spot", and is more _why and _how various labs get the results they do. This is also magnified by the fact that I'm not only interested in hosted providers of inference but local models as well. What's your take on the best model to run for coding on 24GB of VRAM locally after the last few weeks of releases? Which harness do you prefer? What quants do you think are best? To use your sports metaphor it's more than following the national leagues but also following college and even high school leagues as well. And the real interest isn't even who's doing well but WHY, at each level.

nickandbro•1h ago

Very impressive throughput performance

jdeng•1h ago

Excited that the long awaited v4 is finally out. But feel sad that it's not multimodal native.

fblp•1h ago

There's something heartwarming about the developer docs being released before the flashy press release.

onchainintel•1h ago

Insert obligatory "this is the way" Mando scene. Indeed!

necovek•52m ago

Where's the training data and training scripts since you are calling this open source?

Aliabid94•1h ago

MMLU-Pro:

Gemini-3.1-Pro at 91.0

Opus-4.6 at 89.1

GPT-5.4, Kimi2.6, and DS-V4-Pro tied at 87.5

Pretty impressive

ant6n•20m ago

Funny how Gemini is theoretically the best -- but in practice all the bugs in the interface mean I don't want to use it anymore. The worst is it forgets context (and lies about it), but it's very unreliable at reading pdfs (and lies about it). There's also no branch, so once the context is lost/polluted, you have to start projects over and build up the context from scratch again.

KaoruAoiShiho•1h ago

SOTA MRCR (or would've been a few hours earlier... beaten by 5.5), I've long thought of this as the most important non-agentic benchmark, so this is especially impressive. Beats Opus 4.7 here

shafiemoji•1h ago

I hope the update is an improvement. Losing 3.2 would be a real loss, it's excellent.

rvz•1h ago

The paper is here: [0]

Was expecting that the release would be this month [1], since everyone forgot about it and not reading the papers they were releasing and 7 days later here we have it.

One of the key points of this model to look at is the optimization that DeepSeek made with the residual design of the neural network architecture of the LLM, which is manifold-constrained hyper-connections (mHC) which is from this paper [2], which makes this possible to efficiently train it, especially with its hybrid attention mechanism designed for this.

There was not that much discussion around it some months ago here [3] about it but again this is a recommended read of the paper.

I wouldn't trust the benchmarks directly, but would wait for others to try it for themselves to see if it matches the performance of frontier models.

Either way, this is why Anthropic wants to ban open weight models and I cannot wait for the quantized versions to release momentarily.

[0] https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro/blob/main...

[1] https://news.ycombinator.com/item?id=47793880

[2] https://arxiv.org/abs/2512.24880

[3] https://news.ycombinator.com/item?id=46452172

jeswin•1h ago

> this is why Anthropic wants to ban open weight models

Do you have a source?

louiereederson•5m ago

More like he wants to ban accelerator chip sales to China, which may be about “national security” or self preservation against a different model for AI development which also happens to be an existential threat to Anthropic. Maybe those alternatives are actually one and the same to him.

jessepcc•1h ago

At this point 'frontier model release' is a monthly cadence, Kimi 2.6 Claude 4.6 GPT 5.5 — the interesting question is which evals will still be meaningful in 6 months.

swrrt•1h ago

Any visualised benchmark/scoreboard for comparison between latest models? DeepSeek v4 and GPT-5.5 seems to be ground breaking.

raincole•1h ago

History doesn't always repeat itself.

But if it does, then in the following week we'll see DeepSeek4 floods every AI-related online space. Thousands of posts swearing how it's better than the latest models OpenAI/Anthropic/Google have but only costs pennies.

Then a few weeks later it'll be forgotten by most.

sbysb•53m ago

It's difficult because even if the underlying model is very good, not having a pre-built harness like Claude Code makes it very un-sticky for most devs. Even at equal quality, the friction (or at least perceived friction) is higher than the mainstream models.

raincole•47m ago

OpenCode? Pi?

If one finds it difficult to set up OpenCode to use whatever providers they want, I won't call them 'dev'.

The only real friction (if the model is actually as good as SOTA) is to convince your employer to pay for it. But again if it really provides the same value at a fraction of the cost, it'll eventually cease to be an issue.

cmrdporcupine•44m ago

They have instructions right on their page on how to use claude code with it.

ls612•1h ago

How long does it usually take for folks to make smaller distills of these models? I really want to see how this will do when brought down to a size that will run on a Macbook.

inventor7777•41m ago

Weren't there some frameworks recently released to allow Macs to stream weights from fast SSDs and thus fit way more parameters than what would normally fit in RAM?

I have never tried one yet but I am considering trying that for a medium sized model.

the_sleaze_•37m ago

Do you have the links for those? Very interested

inventor7777•32m ago

Sure!

Note: these were just two that I starred when I saw them posted here. I have not looked seriously at it at the moment,

https://github.com/danveloper/flash-moe

https://github.com/t8/hypura

simonw•26m ago

I've been calling that the "streaming experts" trick, the key idea is to take advantage of Mixture of Expert models where only a subset of the weights are used for each round of calculations, then load those weights from SSD into RAM for each round.

As I understand it if DeepSeek v4 Pro is a 1.6T, 49B active that means you'd need just 49B in memory, so ~100GB at 16 bit or ~50GB at 8bit quantized.

v4 Flash is 284B, 13B active so might even fit in <32GB.

inventor7777•22m ago

Ahh, that actually makes more sense now. (As you can tell, I just skimmed through the READMEs and starred "for later".)

My Mac can fit almost 70B (Q3_K_M) in memory at once, so I really need to try this out soon at maybe Q5-ish.

simonw•34m ago

Unsloth often turn them around within a few hours, they might have gone to bed already though!

Keep an eye on https://huggingface.co/unsloth/models

Update ten minutes later: https://huggingface.co/unsloth/DeepSeek-V4-Pro just appeared but doesn't have files in yet, so they are clearly awake and pushing updates.

zargon•1h ago

The Flash version is 284B A13B in mixed FP8 / FP4 and the full native precision weights total approximately 178 GB. KV cache is said to take 10% as much space as V3. This looks very accessible for people running "large" local models. It's a nice follow up to the Gemma 4 and Qwen3.5 small local models.

sbinnee•41m ago

Price is appealing to me. I have been using gemini 3 flash mainly for chat. I may give it a try.

input: $0.14/$0.28 (whereas gemini $0.5/$3)

Does anyone know why output prices have such a big gap?

frozenseven•1h ago

Better link:

https://news.ycombinator.com/item?id=47885014

https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro

reenorap•53m ago

Which version fits in a Mac Studio M3 Ultra 512 GB?

simonw•36m ago

The Flash one should - it's 160GB on Hugging Face: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash/tree/ma...

ycui1986•21m ago

So, dual RTX PRO 6000

sidcool•51m ago

Truly open source coming from China. This is heartwarming. I know if the potential ulterior motives.

I_am_tiberius•34m ago

Open weight!

try-working•21m ago

if you want to understand why labs open source their models: http://try.works/why-chinese-ai-labs-went-open-and-will-rema...

namegulf•45m ago

Is there a Quantized version of this?

yanis_t•38m ago

Already on Openrouter. Pro version is $1.74/m/input, $3.48/m/output, while flash $0.14/m/input, 0.28/m/output.

esafak•27m ago

https://openrouter.ai/deepseek/deepseek-v4-pro

https://openrouter.ai/deepseek/deepseek-v4-flash

77ko•17m ago

Its on OR - but currently not available on their anthropic endpoint. OR if you read this, pls enable it there! I am using kimi-2.6 with Claude Code, works well, but Deepseek V4 gives an error:

`https://openrouter.ai/api/messages with model=deepseek/deepseek-v4-pro, OR returns an error because their Anthropic-compat translator doesn't cover V4 yet. The Claude CLI dutifully surfaces that error as "model...does not exist"

astrod•18m ago

Getting 'Api Error' here :( Every other model is working fine.

aliljet•37m ago

How can you reasonably try to get near frontier (even at all tps) on hardware you own? Maybe under 5k in cost?

awakeasleep•24m ago

The same way you fit a bucket wheel excavator in your garage

jdoe1337halo•22m ago

More like 500k

542458•6m ago

The low end could be something like an eBay-sourced server with a truckload of DDR3 ram doing all-cpu inference - models with a terabyte of ram can be had for about 1.5K. The TPS will be absolute garbage, but it will nominally run.

hongbo_zhang•37m ago

congrats

simonw•37m ago

I like the pelican I got out of deepseek-v4-flash more than the one I got from deepseek-v4-pro.

Flash: https://gist.github.com/simonw/4a7a9e75b666a58a0cf81495acddf...

Pro: https://gist.github.com/simonw/9e8dfed68933ab752c9cf27a03250...

Both generated using OpenRouter.

For comparison, here's what I got from DeepSeek 3.2 back in December: https://simonwillison.net/2025/Dec/1/deepseek-v32/

And DeepSeek 3.1 in August: https://simonwillison.net/2025/Aug/22/deepseek-31/

And DeepSeek v3-0324 in March last year: https://simonwillison.net/2025/Mar/24/deepseek/

JSR_FDED•35m ago

No way. The Pro pelican is fatter, has a customized front fork, and the sun is shining! He’s definitely living the best life.

w4yai•31m ago

yeah. look at these 4 feathers (?) on his bum too.

oliver236•12m ago

a lot of dumplings

nickvec•32m ago

The Flash one is pretty impressive. Might be my favorite so far in the pelican-riding-a-bicycle series

ycui1986•26m ago

I really like the pro version. The pelican is so cute.

whateveracct•5m ago

i wish ppl would shut up about the pelican. it always has this scientific air which is so wrong.

mchusma•29m ago

For comparison on openrouter DeepSeek v4 Flash is slightly cheaper than Gemma 4 31b, more expensive than Gemma 4 26b, but it does support prompt caching, which means for some applications it will be the cheapest. Excited to see how it compares with Gemma 4.

mariopt•24m ago

Does deepseek has any coding plan?

jeffzys8•5m ago

dhruv3006•14m ago

Ah now !

FujiNet Go 800 – Atari800 Emulator for Android

The Surveillance Accountability Act Full Text [pdf]

OpenAI deprecates all GPT nano fine tuning

Why Not Venus?

Running Bare-Metal Rust Alongside ESP-IDF on the ESP32-S3's Second Core

The Budgeting Mistake That Cost Uber Its Annual AI Spend in 4 Months

Tremendous Iranian Invasion: A Text Misadventure

Essential Voice by Nothing

Familiarity is the enemy: On why Enterprise systems have failed for 60 years

Intel Arc Pro B70 Review

ASML's latest chipmaking gear is too pricey, even for TSMC

Intel Arc Pro B70 benchmarks for LLMs and video generation

DeepSeek's Sequel Set to Extend China's Reach in Open-Source A.I

Ubuntu 26.04 LTS Released

AI Resume Reviewer

Show HN: GitRails-Let agents call only the GitHub endpoints and params you allow

The AI Science Separation

We're Using So Much AI That Computing Firepower Is Running Out

Grok Voice Think Fast 1.0

Oracle's Deluge of AI Debt Pushes Wall Street to the Limit

Canonical Releases Ubuntu 26.04 LTS Resolute Raccoon

Unlimited access to Reddit, and Twitter, access anything with supermcp

Anthropic tested removing Claude Code from the Pro plan

S&box – Game engine built on Source 2 and .NET from creators of Garry's Mod

Share X 20.0

Habitual coffee intake shapes the microbiome, modifies physiology and cognition

New Blood CEO Dave Oshry on Indie Strategy: "We Pump New Blood into Old Genres"

Shelter Aid

Mullvad to add feature that forces all iOS traffic through the VPN tunnel

AI run store in SF can't stop ordering candies and paying women less.