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

DeepSeek V4 Flash 0731

arcprize.org

Post reply on HN
Print view View original

211–220 of 477 posts

Re: DeepSeek V4 Flash 0731

#211

Post by apitman »

ak_t wrote: post #164

Note this is the 07/31 release of DSv4 flash and not the "preview" that they put out a couple months or so ago. I've been running this model locally for a week, and the preview version before that. This updated one feels like a whole tier up. It's very capable for debugging and analyzing documents/data I upload. The killer feature, IMO, is the speed. On 2x RTX Pro 6000 Blackwell, its ~8k tok/s prefill and ~250 tok/s…

I'm getting like 25 tok/s on 2x RTX Pro 6000. This is with llama.cpp, but I had GPT tune it for me. I was under the impression vLLM was at most ~2x faster, and usually for highly parallel loads. Any tips on where I should look first for an obvious blunder?

I'm guessing tensor parallelism or similar?

Re: DeepSeek V4 Flash 0731

#212

Post by kzrdude »

ignoramous wrote: post #155
NoboruWataya wrote: post #138

My Claude account was banned the other day. The only possible cause I can think of is that I tried to authenticate from the AI assistant in a JetBrains IDE and, not thinking, entered the details for my regular subscription rather than an API account. As soon as it became apparent that I needed an API account rather than a subscription, I just closed out of the tab. Nevertheless, about 20 minutes later I got an email…

> My initial thought was to sign up for ChatGPT, but I had $20 in OpenRouter so I've been trying out DeepSeek V4 Pro with Pi for the last few days and I gotta say, it's good enough for my use case If you prefer subscriptions, OpenCode Go ($10/mo), Cline Pass ($10/mo), Atlas Code ($20/mo), and CommandCode ($1/mo) serve some of the best open weights with generous limits. OpenCode Go currently offers $120 for $10 on Dee…

Just trying to understand, https://opencode.ai/docs/go/#privacy currently says DeepSeek V4 Flash has 0 days data retention.

> DeepSeek V4 Flash: ZDR agreement is renewed monthly. The current agreement is valid through August 31, 2026.

Is there other info I should be aware of w.r.t data retention with opencode go? It's hosted in China, so other middlemen may be active (I doubt it, but possible)?

Re: DeepSeek V4 Flash 0731

#213

Post by namr2000 »

ak_t wrote: post #164

Note this is the 07/31 release of DSv4 flash and not the "preview" that they put out a couple months or so ago. I've been running this model locally for a week, and the preview version before that. This updated one feels like a whole tier up. It's very capable for debugging and analyzing documents/data I upload. The killer feature, IMO, is the speed. On 2x RTX Pro 6000 Blackwell, its ~8k tok/s prefill and ~250 tok/s…

What runtime are you using with the 2x RTX Pro 6000 Blackwell machine? I have the same setup and tried DSv4 Flash on vLLM and ran into a ton of kernel bugs that don't seem to have been fixed yet.

Re: DeepSeek V4 Flash 0731

#214

Post by natrys »

greenavocado wrote: post #106
akman wrote: post #39

Earlier quoted context omitted.

90%+ cache hit rate is common, and so you'll see on places like openrouter that Deepseek cache cost is indeed a magnitude cheaper than the rest.

My usage thus far from api.deepseek.com - input_cache_hit_tokens: 1,265,646,976 x 0.0000000028 = $3.5438115328 - input_cache_miss_tokens: 18,208,088 x 0.00000014 = $2.54913232 - output_tokens: 9,615,178 x 0.00000028 = $2.69224984 - request_count: 10,837 (no price) Total cost: $8.7851936928 (approximately $8.79) Cache: - Hit: 1,265,646,976 - Miss: 18,208,088 - Total input tokens: 1,283,855,064 Hit rate: 98.582% (1,265…

So your cost would be $40.68 with another provider that has one less zero in the cache hit price.

Re: DeepSeek V4 Flash 0731

#215

Post by zargon »

ComputerGuru wrote: post #171
ak_t wrote: post #164

Note this is the 07/31 release of DSv4 flash and not the "preview" that they put out a couple months or so ago. I've been running this model locally for a week, and the preview version before that. This updated one feels like a whole tier up. It's very capable for debugging and analyzing documents/data I upload. The killer feature, IMO, is the speed. On 2x RTX Pro 6000 Blackwell, its ~8k tok/s prefill and ~250 tok/s…

What quantization level is that? Because official endpoints are slow .

V4 Flash fits entirely in two RTX Pro 6000s without any quantization at all.

Re: DeepSeek V4 Flash 0731

#216

Post by WarmWash »

kzrdude wrote: post #212
ignoramous wrote: post #155

Earlier quoted context omitted.

> My initial thought was to sign up for ChatGPT, but I had $20 in OpenRouter so I've been trying out DeepSeek V4 Pro with Pi for the last few days and I gotta say, it's good enough for my use case If you prefer subscriptions, OpenCode Go ($10/mo), Cline Pass ($10/mo), Atlas Code ($20/mo), and CommandCode ($1/mo) serve some of the best open weights with generous limits. OpenCode Go currently offers $120 for $10 on Dee…

Just trying to understand, https://opencode.ai/docs/go/#privacy currently says DeepSeek V4 Flash has 0 days data retention. > DeepSeek V4 Flash: ZDR agreement is renewed monthly. The current agreement is valid through August 31, 2026. Is there other info I should be aware of w.r.t data retention with opencode go? It's hosted in China, so other middlemen may be active (I doubt it, but possible)?

If it's hosted in China, they can tell you whatever you want to hear and do whatever they want to do.

What are you going to do? Take a CCP company in front of a CCP judge?

Re: DeepSeek V4 Flash 0731

#217

Post by natrys »

nylonstrung wrote: post #173

Compared to the last Deepseek V4 Flash version I've had tons of issues with it getting in infinite loops and talking to itself without executing tool calls, wasting tons of tokens This is on Pi agent, nothing fancy at all about my prompts or use case. Anyone else experiencing this? I've also had it randomly go from talking about Rust to talking about the electric chair, controversies about D&D rules (both irrelevant…

I think this one requires a bit of strong prompting. I am also normally a Pi user, but my experience in OpenCode with this model has been drastically better than in Pi, where it overthinks a lot and gets distracted by random things.

It might be even better in Codex or Oh My Pi according to this bench I saw earlier: https://nitter.net/composio/status/2085330847951970801

Re: DeepSeek V4 Flash 0731

#218

Post by zmmmmm »

it's great but we need a multi-modal model of this quality and price to truly declare victory.

But it makes me quite curious, how a text-only model can do so well on ARC-AGI-2 being a set of visual puzzles? It would have to solve it entirely using text-only spatial reasoning about the grid (or maybe writing code?). I am curious if this is normal or do other models use their vision capabilities to solve the puzzles?

Re: DeepSeek V4 Flash 0731

#219

Post by seanmcdirmid »

Note they double the price if you use during peak time. However, they define peak time with respect to China, not Europe or the USA...so if you are out of Asia, I guess Australians might be impacted, and its still cheap anyways.

Re: DeepSeek V4 Flash 0731

#220

Post by LaurensBER »

paxys wrote: post #118
LaurensBER wrote: post #9

I've been using it extensively since the release and the best summary I can give is that it's good enough to use it for (almost) everything and cheap enough that the cost are irrelevant. I'm running it in Oh My Pi with a second instance running as "advisor" and even with 5-6 active sessions (effectively 12 streams) I'm struggling to spend more than 5 bucks per day. OpenCode Go even has double limits temporarily so fo…

How is $5/day irrelevant? In the $150/mo range you can get effectively unlimited usage of GPT 5.6 Sol (Pro plan). Why use a much weaker model for the same price?

5 USD is at the "raw" API price.

OpenCode currently offers 60 USD API credits at 10 USD per month (OpenCode Go) and have even doubled it temporarily as a promotion.

Effectively you can get Deepseek for 1/12th the already ridiculous cheap API price.

Post reply on HN