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

DeepSeek-V4-Flash Update

api-docs.deepseek.com

Post reply on HN
Print view View original

51–60 of 362 posts

Re: DeepSeek-V4-Flash Update

#51

Post by bayesianbot »

throwaw12 wrote: post #15
dnhkng wrote: post #2

DeepSeek V4 Flash (Preview → 2026-07-31) • Terminal Bench: 56.9 → 82.7 (+25.8) • Toolathlon: 51.8 → 70.3 (+18.5) Compared to GPT-5.6 Terra: • Terminal Bench: Flash 82.7 vs Terra 78.4 • Toolathlon: Flash 70.3 vs Terra 53.1 • DeepSWE: Flash 54.4 vs Terra 69.6 • Agents' Last Exam: Flash 25.2 vs Terra 50.4 Trading blows with Terra, which is pretty interesting. No clear winner on these benchmarks, and wildy differeing sco…

Open flash model is competing against OpenAI's 'Sonnet' model at the price of GPT 3, I am really excited about this release, hopefully it holds up in real work as well

IIRC GPT 3 was priced at per 1k tokens, had to check, the biggest GPT 3 model from OpenAI was $0.06/1k, so $60 / 1M. gpt-3.5-turbo was the first model after ChatGPT and that was $2 / 1M. And no caching. So not really in the same ballpark

Re: DeepSeek-V4-Flash Update

#52

Post by amunozo »

storywatch wrote: post #37

How's their performance in English prose? We are currently searching for cost effective ways to keep story wikis up to date.

I am interested in this too, as I think it could be these models are overly optimized for coding. Let me know if you figure it out!

Re: DeepSeek-V4-Flash Update

#53

Post by kmarc »

Essentially I'm running everything on flash now inside pi. With the correct set of MCP servers, context reducer tooling and skills it can implement any task I throw at it. Some sessions take 30+ turns, but it's fast and cheap; all this in an hour, with ~$0.5 cost.

(TBH though, in my multi-subagent workflow I do use other, more expensive models for planning, reviewing, oracle-ing)

I haven't used our slow opus subscription for weeks.

(Also set up an OpenWebUi self-hosted chat that works from my phone, has some mcp and skills. fully replaced perplexity. Monthly cost ~$18 for hosting and subscriptions)

Re: DeepSeek-V4-Flash Update

#54

Post by petu »

PhilippGille wrote: post #30

The previous V4 version wasn't called “Preview” by most inference providers. For example, the OpenRouter model slug was `deepseek/deepseek-v4-flash`. So now there will be confusion when someone talks about V4 Flash or when someone offers V4 Flash inference. Why not call it V4.1?

It probably would be called 'deepseek-v4-flash-0731' in API

edit: nope, at least deepseek kept "deepseek-v4-flash" and just updated model underneath. I guess preview is no longer worth serving with that release and you'd have to look through inference provider docs to see if they've updated, yeah..

Re: DeepSeek-V4-Flash Update

#55

Post by peperunas »

kmarc wrote: post #53

Essentially I'm running everything on flash now inside pi. With the correct set of MCP servers, context reducer tooling and skills it can implement any task I throw at it. Some sessions take 30+ turns, but it's fast and cheap; all this in an hour, with ~$0.5 cost. (TBH though, in my multi-subagent workflow I do use other, more expensive models for planning, reviewing, oracle-ing) I haven't used our slow opus subscrip…

Do you have any recommendations of such extensions for pi?

Re: DeepSeek-V4-Flash Update

#56

Post by embedding-shape »

f311a wrote: post #38

I've been driving flash model for 90% of my tasks. It's better than pro (for unknown reasons), very cheap and fast. I try to keep changes under 1000 lines and drive architectural decisions myself, barely notice any difference compared to frontier models. The rest 10% is to spot bugs, security problems and to investigate better architecture, which flash can also do pretty well, I just cross check it. Faster iterations…

> Also, it will never complain about security guards, I've been using it to reverse engineer binaries.

Maybe I'm using too weak language in my prompts, but none of the OpenAI models I've used via codex has refused to reverse engineer binaries, is it supposed to? I'm sitting right now reverse-engineering a 3rd party firmware together with Codex and haven't hit a single guardrail. Meanwhile, I see people complaining about it rejecting non-security related prompts, are things so individual on the platforms right now or what's going on?

Re: DeepSeek-V4-Flash Update

#57

Post by vladukha »

Where do you guys get deepseek? I'm hearing a lot of good reviews and want to try it with my pi config. from the deeepseek themselves, openrouter, or anywhere else? does it make a difference? [edit]: whoa it is really fast. will take some time to evaluate quality thou

Re: DeepSeek-V4-Flash Update

#58

Post by matrik »

NitpickLawyer wrote: post #43
lukan wrote: post #39

Earlier quoted context omitted.

"it'll barely run on an M5 Max " The max version I could order now with 128 GB? If so, the price for local inference would be 12 000 € vs 500 000 € for a B300.

There's also the 2x spark way, which should be ~8k eur? Someone down the thread reported ~60tps for 2x sparks. That's totally usable for local inference. You can also do 2x 6kPRO in a workstation, for ~20k.

For the same performance, one could even go about 50% cheaper with 16 channel ddr4 + a rtx3090 for prompt processing.

But still, even for mid level projects API is orders of magnitude cheaper, since you don't need to set it up and maintain it.

Re: DeepSeek-V4-Flash Update

#59

Post by reverius42 »

lukan wrote: post #39
spwa4 wrote: post #4

In case people want to run it, it's DeepSeek-V4-Flash-284B-A13B. So it should just barely run on a single B300, and it's small enough that it'll barely run on an M5 Max too.

"it'll barely run on an M5 Max " The max version I could order now with 128 GB? If so, the price for local inference would be 12 000 € vs 500 000 € for a B300.

I'm running a useful quantization of the previous version of Deepseek-V4-Flash -- quite well but with so much fan noise -- on a MacBook Pro M5 Max with 128 GB.

Re: DeepSeek-V4-Flash Update

#60

Post by spwa4 »

NitpickLawyer wrote: post #43
lukan wrote: post #39

Earlier quoted context omitted.

"it'll barely run on an M5 Max " The max version I could order now with 128 GB? If so, the price for local inference would be 12 000 € vs 500 000 € for a B300.

There's also the 2x spark way, which should be ~8k eur? Someone down the thread reported ~60tps for 2x sparks. That's totally usable for local inference. You can also do 2x 6kPRO in a workstation, for ~20k.

Currently the 3bit (and 2 bit) quant on DGX spark (on one of them) and the M5 Max should just start. Right now. (I'm hoping to get an M5 Max delivered on monday, let's see if it happens this time. It's 2+ months since I ordered now)

The 4 bit quant technically fits (there's a 127 GB version) but ... obviously that's not going to work. It is so close though, surely someone will a way to do it.

Post reply on HN