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

Cerebras launches Qwen3-235B, achieving 1.5k tokens per second

cerebras.ai

Post reply on HN
Print view View original

101–110 of 160 posts

Re: Cerebras launches Qwen3-235B, achieving 1.5k tokens per second

#101

Post by meowface »

throwaw12 wrote: post #31
meowface wrote: post #20

Earlier quoted context omitted.

I just set up Groq with Kimi K2 the other day and was blown away by the speed. Deciding if I should switch to Qwen 3 and Cerebras. (Also, off-topic, but the name reminds me of cerebrates from Starcraft. The Zerg command hierarchy lore was fascinating when I was a young child.)

Have you used Claude Code and how do you compare the quality to Claude models? I am heavily invested in tools around Claude, still struggling to make a switch and start experimenting with other models

I still exclusively use Claude Code. I have not yet experimented with these other models for practical software development work.

A workflow I've been hearing about is: use Claude Code until quota exhaustion, then use Gemini CLI with Gemini 2.5 Pro free credits until quota exhaustion, then use something like a cheap-ish K2 or Qwen 3 provider, with OpenCode or the new Qwen Code, until your Claude Code credits reset and you begin the cycle anew.

Re: Cerebras launches Qwen3-235B, achieving 1.5k tokens per second

#102

Post by aurareturn »

sailingparrot wrote: post #37
aurareturn wrote: post #19

If this is the full fp16 quant, you'd need 2TB of memory to use with the full 131k context. With 44GB of SRAM per Cerebras chip, you'd need 45 chips chained together. $3m per chip. $135m total to run this. For comparison, you can buy a DGX B200 with 8x B200 Blackwell chips and 1.4TB of memory for around $500k. Two systems would give you 2.8TB memory which is enough for this. So $1m vs $135m to run this model. It's no…

> With 44GB of SRAM per Cerebras chip, you'd need 45 chips chained together. $3m per chip. $135m total to run this. That's not how you would do it with Cerebras. 44GB is SRAM, so on chip memory, not HBM memory where you would store most of the params. For reference one GB200 has only 126MB of SRAM, if you tried to estimate how many GB200 you would need for a 2TB model just by looking at the L2 cache size you would ge…

  That's not how you would do it with Cerebras. 44GB is SRAM, so on chip memory, not HBM memory where you would store most of the params. For reference one GB200 has only 126MB of SRAM, if you tried to estimate how many GB200 you would need for a 2TB model just by looking at the L2 cache size you would get 16k GB200 aka ~600M$, obviously way off.
Yes but Cerebras achieves its speed by using SRAM.

Re: Cerebras launches Qwen3-235B, achieving 1.5k tokens per second

#103

Post by npsomaratna »

Adfeldman wrote: post #99
aurareturn wrote: post #19

If this is the full fp16 quant, you'd need 2TB of memory to use with the full 131k context. With 44GB of SRAM per Cerebras chip, you'd need 45 chips chained together. $3m per chip. $135m total to run this. For comparison, you can buy a DGX B200 with 8x B200 Blackwell chips and 1.4TB of memory for around $500k. Two systems would give you 2.8TB memory which is enough for this. So $1m vs $135m to run this model. It's no…

Our chips don't cost $3M. I'm not sure where you got that number but its wildly incorrect.

Are you the CEO of Cerebras? (Guessing from the handle)

Re: Cerebras launches Qwen3-235B, achieving 1.5k tokens per second

#104

Post by skeezyboy »

lazide wrote: post #100
skeezyboy wrote: post #87

Earlier quoted context omitted.

>But then, same for humans yes? And? Whats your point? This is a computer. Humans make errors doing arithmetic, therefore should we not expect computers to be able to reliably perform arithmetic? No. Silly retort and a common reply from people who are suitably wowed by the current generation of AI.

This is incredibly dumb.

whats what im trying to tell you

Re: Cerebras launches Qwen3-235B, achieving 1.5k tokens per second

#105

Post by sailingparrot »

aurareturn wrote: post #102
sailingparrot wrote: post #37

Earlier quoted context omitted.

> With 44GB of SRAM per Cerebras chip, you'd need 45 chips chained together. $3m per chip. $135m total to run this. That's not how you would do it with Cerebras. 44GB is SRAM, so on chip memory, not HBM memory where you would store most of the params. For reference one GB200 has only 126MB of SRAM, if you tried to estimate how many GB200 you would need for a 2TB model just by looking at the L2 cache size you would ge…

That's not how you would do it with Cerebras. 44GB is SRAM, so on chip memory, not HBM memory where you would store most of the params. For reference one GB200 has only 126MB of SRAM, if you tried to estimate how many GB200 you would need for a 2TB model just by looking at the L2 cache size you would get 16k GB200 aka ~600M$, obviously way off. Yes but Cerebras achieves its speed by using SRAM.

There is no way not to use SRAM on a GPU/Cerebras/most accelerators. This is where the cores fetch the data.

But that doesn’t mean you are only using SRAM, that would be impractical. Just like using a CPU just by storing stuff in the L3 cache and never going to the RAM. Unless I am missing something from the original link, I don’t know how you got to the conclusion that they only used SRAM.

Re: Cerebras launches Qwen3-235B, achieving 1.5k tokens per second

#107

Post by htrp »

aurareturn wrote: post #19

If this is the full fp16 quant, you'd need 2TB of memory to use with the full 131k context. With 44GB of SRAM per Cerebras chip, you'd need 45 chips chained together. $3m per chip. $135m total to run this. For comparison, you can buy a DGX B200 with 8x B200 Blackwell chips and 1.4TB of memory for around $500k. Two systems would give you 2.8TB memory which is enough for this. So $1m vs $135m to run this model. It's no…

>Maybe hedge funds or some sort of financial markets?

Definitely not hedge funds / quant funds.

You'd just buy a dgx

Re: Cerebras launches Qwen3-235B, achieving 1.5k tokens per second

#108

Post by aurareturn »

sailingparrot wrote: post #105
aurareturn wrote: post #102

Earlier quoted context omitted.

That's not how you would do it with Cerebras. 44GB is SRAM, so on chip memory, not HBM memory where you would store most of the params. For reference one GB200 has only 126MB of SRAM, if you tried to estimate how many GB200 you would need for a 2TB model just by looking at the L2 cache size you would get 16k GB200 aka ~600M$, obviously way off. Yes but Cerebras achieves its speed by using SRAM.

There is no way not to use SRAM on a GPU/Cerebras/most accelerators. This is where the cores fetch the data. But that doesn’t mean you are only using SRAM, that would be impractical. Just like using a CPU just by storing stuff in the L3 cache and never going to the RAM. Unless I am missing something from the original link, I don’t know how you got to the conclusion that they only used SRAM.

I know Groq chips load the entire model into SRAM. That's why it can be so fast.

So if Cerebras uses HBM to store the model but stream weights into SRAM, I really don't see the advantage long term over smaller chips like GB200 since both architectures use HBM.

The whole point of having a wafer chip is that you limit the need to reach out to external parts for memory since that's the slow part.

Re: Cerebras launches Qwen3-235B, achieving 1.5k tokens per second

#109

Post by aurareturn »

Adfeldman wrote: post #99
aurareturn wrote: post #19

If this is the full fp16 quant, you'd need 2TB of memory to use with the full 131k context. With 44GB of SRAM per Cerebras chip, you'd need 45 chips chained together. $3m per chip. $135m total to run this. For comparison, you can buy a DGX B200 with 8x B200 Blackwell chips and 1.4TB of memory for around $500k. Two systems would give you 2.8TB memory which is enough for this. So $1m vs $135m to run this model. It's no…

Our chips don't cost $3M. I'm not sure where you got that number but its wildly incorrect.

So how much does it cost? Google search return $3m. Here's your chance to tell us your real price if you disagree.

Re: Cerebras launches Qwen3-235B, achieving 1.5k tokens per second

#110

Post by qualeed »

Adfeldman wrote: post #99
aurareturn wrote: post #19

If this is the full fp16 quant, you'd need 2TB of memory to use with the full 131k context. With 44GB of SRAM per Cerebras chip, you'd need 45 chips chained together. $3m per chip. $135m total to run this. For comparison, you can buy a DGX B200 with 8x B200 Blackwell chips and 1.4TB of memory for around $500k. Two systems would give you 2.8TB memory which is enough for this. So $1m vs $135m to run this model. It's no…

Our chips don't cost $3M. I'm not sure where you got that number but its wildly incorrect.

In that case, mind providing a more appropriate ballpark?
Post reply on HN