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

91–100 of 215 posts

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

#91

Post by m00dy »

ricardobeat wrote: post #24
darkoob12 wrote: post #23

Earlier quoted context omitted.

My mental bias always kept me away from Chinese models. Because i know that china is a surveillance state and all the things we know about CCP. But after what we learned about OpenAI and how they most likely used user data to basically cheat in an open competition i think it does not matter which AI provider you use all of them will own your data and all of them can spy on you. So I am willing to switch to Chinese mo…

These models are open-weights. Anyone can host them, you don’t have to use chinese servers even though most of them offer zero data-retention policies.

>>zero data-retention policies

Yeah, that’s basically an industry-wide scam.

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

#92

Post by gentlewater »

swiftcoder wrote: post #22
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 hit this too, but you can just add "in English" to steer it

I finally uninstalled the app yesterday after giving it plenty of chances over several months. Yesterday, I asked it whether «DeepSeek has fixed the issue where it erroneously answers in Chinese?» and it answered in Chinese.

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

#93

Post by pimeys »

mattmcal wrote: post #57
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.

All of these flash models have this. You have to build your harness so that it deals with it. Infinite loops are solved by having an error message that says what to do differently on failure, invalid tool calls are solved by making the tool schema less strict and detect things in the runtime etc.

Hallucinations you can't fix. Gemini is a bit worse there than DeepSeek, but there's not much research on how to fix that. The only one is the CaMeL paper by Google, where you tag every prompt and result and then for every assistant response or tool call you first check where it got that data and error if you notice fabrication. This one is really annoying to implement.

With larger models the fabrication starts when the context grows or if you have too many tools, for flash models it's much earlier. We use the flash models for repetitive agentic tasks, where the prompt defines clearly what to do and how. The whole run is about 4-5 steps typically, and context size stays in the comfort zone.

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

#94

Post by KoolKat23 »

aftbit wrote: post #67

>In keeping with our commitment to user responsibility, following the official launch of V4.1 Flash and prior to the release of V4.1 Pro, all requests to the Pro model will be routed to V4.1 Flash and billed at Flash's price Please don't do this kind of thing. If a user has validated a workflow on V4 Pro, they might not want to suddenly start testing it in production on V4.1 Flash. Instead, keep V4 Pro around but dep…

I imagine they need the compute. Can expand market share with more users for same amount of compute.

But I agree with you. I have a dumb workflow that worked well with v4-flash-0731 and I suspect is directing to a newer model that now breaks it.

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

#95

Post by tarruda »

fluoridation wrote: post #40
tarruda wrote: post #18

Earlier quoted context omitted.

IQ3_XXS (~3.2 BPW). For me this is an option because my Mac studio is only used for serving LLMs, so I can afford to dedicate most of its RAM to this. I can run with 256k context and only uses ~117G, with the remaining (up to 125G which I can allocate to VRAM) being used for prompt caching and context checkpoints. I'm making my own quants, though the Vision-Exp version is outdated and won't work on llama.cpp master b…

Thanks, I'll give that a try. I basically have the same use case, only on Strix Halo.

Don't use my Vision-Exp GGUF though. As I said I built those GGUFs before llama.cpp supported, and they can't be loaded on current master (require my own branch).

I already have new GGUFs but haven't uploaded yet. If you want Vision-Exp, maybe use bartowski or unsloth's GGUFs.

Side note:

As an alternative to deepseek v4, you might want to give it a shot at qwen 3.8 flash next. I have IQ4_NL GGUFs that can be loaded fully into 128G, or Q5_K GGUFs that can offload the PLE to disk (use --load-mode none --lazy-mode on for that): https://huggingface.co/tarruda/Qwen3.8-Flash-Next-GGUF.

llama.cpp master is still somewhat bad in Qwen 3.8 next performance, but I was able to achieve 40tps tg and 600 tps pp on my private branch.

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

#96

Post by urieiejr »

pimeys wrote: post #37
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.

Yes. I'm working in the agent industry and my god are we excited on new versions of Chinese flash models. The direct competition is Gemini Flash, and these models are much better on agentic tasks with fraction of the task price compared to Gemini. Things like oh here's a set of simple instructions for you to follow, call these tools, return this report. 20-30% of the price per task. And especially Deepseek Flash prod…

translation

I make vaporware that doesnt do shit reliably and this chinese crap spouts plausible demos and spam calls more cheaply than the competition saaar

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

#97

Post by wg0 »

DeepSeek v4 Flash with high is already a really great work horse. Reliable. But this time, not only that it is better but they are reducing the price by 50% so that's great.

I also find the DeepSeek models to be more precise than Claude models (last I used 4.7) in that I yet had not the occasion where model did something unintentional that I did not direct it to.

EDIT: Updated percentage reduction.

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

#98

Post by k__ »

tjwebbnorfolk wrote: post #51
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.

At least they can use another provider or self-host.

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

#99

Post by kennywinker »

wg0 wrote: post #97

DeepSeek v4 Flash with high is already a really great work horse. Reliable. But this time, not only that it is better but they are reducing the price by 50% so that's great. I also find the DeepSeek models to be more precise than Claude models (last I used 4.7) in that I yet had not the occasion where model did something unintentional that I did not direct it to. EDIT: Updated percentage reduction.

Reducing the price by 100% means it’s free. I think you mean reducing the price by 50%

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

#100

Post by kamranjon »

tarruda wrote: post #18
fluoridation wrote: post #10

Earlier quoted context omitted.

Interesting, I had assumed it'd be too large to fit. What quant and context size are you running?

IQ3_XXS (~3.2 BPW). For me this is an option because my Mac studio is only used for serving LLMs, so I can afford to dedicate most of its RAM to this. I can run with 256k context and only uses ~117G, with the remaining (up to 125G which I can allocate to VRAM) being used for prompt caching and context checkpoints. I'm making my own quants, though the Vision-Exp version is outdated and won't work on llama.cpp master b…

Hey there! I do the same but I use dwarfstar at a 2-bit quant: https://github.com/antirez/ds4

I'm curious if you've tried dwarfstar and decided to move to llama.cpp and 3 bit quants or what made you go that route instead? I've been using ds4 for months now and it's already got support for the new vision model, haven't tried it yet, still on 0731 but it's been very solid for me.

Post reply on HN